Tuesday, August 30, 2011

Shooting Stars

Anyone who has followed my Twitter feed over the past few months knows I am big baseball fan, specifically with regards to the statistical side of the game. I am also a big proponent of trying to use lessons from one sport to help us understand or improve another (the Premier League’s stubbornness in ignoring the success stories of officials, salary caps, transfer rules etc from other sports is a huge source on frustration for me). The statistical awareness in baseball is possibly the best of any sport in the world, with legions of fans (and team employees) dreaming up new ways to measure players and hopefully identify undervalued assets and commodities (the now over, and often incorrectly, used term ‘moneyball’ attempts to convey this thought process).

One term you will quickly come across if you ever decided to visit a statistically leaning site like FanGraphs, would be ‘regression to the mean’. I am no statistician so I apologize to anyone that is if the below is a bit simplistic, but, essentially, this means that over a period of time statistical anomalies will level out and a player will produce at a level in line with his talent and/or past performance. This doesn't mean that his future production will account for any past sub-par performance, but it will at least appear to improve to get back to the level he (or the league) usually performs.

Two things to note here. One, it is important to remember that while regression sounds like a negative term, it can go both ways. If a basketball player has a career free throw shooting percentage of 91% but starts a season 21 for his first 28 shots (75%), we would expect, over a longer period, that all else being equal, the average will regress back to his 91% level and so he should enjoy a period of shooting above his career average as things level off. Second, we are dealing with human beings and not robots so other factors do indeed play a role. A player might be unhappy with his contract, finding it hard to settle in a new rainy city (not looking at anyone in particular Carlos) or playing with an undisclosed injury, all of which will have some impact on his play. However, research done in the excellent Scorecasting  suggests that ‘hot streaks’ are generally statistical quirks remembered due to our own desire to explain statistical fluctuations, rather than the result of a player being ‘locked in’.

If you've made it this far, we can get into the actual meat of the post. Taking all forwards who scored more than 5 goals last year, we see that, on average a player took 6.1 shots per goal with 2.9 of them on target. The correlation between total shots and goals is 89% while shots on target is even better at 91%. Taking shots on target a step further, we see that the standard deviation of the sample is just 0.78, suggesting that the shots/goal ratio holds up across the board. Indeed, of the 34 players to score 5 goals or more, 22 scored a goal every 2-3 shots on target and only three players (Drogba, Rodallega and Lovenkrands) had more than 4 on target per goal (Hernandez was the biggest outlier in the other direction, taking just 1.8 shots on target respectively to score).

So if we accept that a high number of shots on target is going to lead to a lot of goals (it sounds obvious but it isn't 100% correlated) then the idea is that we can look at the shot totals to date and if a player is scoring well below his career average rate, we could imply some regression coming in the future (the same goes for a player scoring every time he shoots).

The season so far
A few players stand out from the early going, though be careful here as we are in seriously small sample size range. This review will be more useful after a few more games, but it's worth taking a quick look now:

NameGoalsTotal ShotsOn TargetOT/GCareer Rate
van Persie 

While Dzeko and Rooney's goal rate looks unsustainable (and of course, Dzeko won't score twice a game all year) if they keep having the number of shots they've managed to date, they are going to score some seriously high totals. 10 shots a piece on target through 3 games is off the charts and not even the trigger happy Ronaldo could match those totals. This suggests that Dzeko deserves as much, or more hype than Aguero given his lower price tag and while Tevez is obviously there as a constant threat, Dzeko is not simply tapping in the good work of others.

You'd like to see Aguero and Suarez up there shots on target per game as they surely can't continue to basically score almost every time they hit the target. This pair are two of the most promising players in the league but this evidence does put at least a small blot on their developing resumes.

Though he hasn't racked up big totals on target yet, Gyan's 10 total shots stand out among those players without a goal and if he continues at that clip he will surely finish with a good haul come year end. Andy Carroll hasn't been well received this year but he too, with 9 shots in a couple of games, could be in line for some adjustment over the coming weeks.

For the future
As I say, this is more something to re-examine in GW8 or so but it is worth looking at nonetheless. The extreme totals produced by Rooney and Dzeko look like they could be legit and Dzeko is very much on my radar alongside Aguero for the coming weeks. We will check back on these totals in a few weeks.

I now have a full set of stats for the new season to work with, so keep checking back throughout the international break as we try and find some clues as to where this crazy season fantasy season is heading next.

Monday, August 29, 2011

Reader Request

A quick reader request. For reasons unknown, the powers at Premier League towers have decided to remove the match data from their once reliable website instead opting for an unordered list of random facts including goalscorers, yellow cards etc. To calculate the various data I use on this site I need a listing of player stats each week including minutes played, goals, assists, bonus points and fantasy points and then I need that cross referenced by opponent to establish where (home/away) and against whom (good /bad teams) a particular player is scoring their points.

I can get goal and assist data from ESPN, who provide full match data but they obviously do not include fantasy or bonus points and annoyingly use full names rather than surnames, which requires some excel shenanigans just to get into a readable state.

Does anyone (a) know of any sources which list match data including minutes played or (b) have a reliably maintained spreadsheet with the data included?

Sunday, August 28, 2011

Lineup Lessons - Gameweek 3

Szczesny, Jenkinson, Djourou, Koscielny, Traore, Ramsey, Coquelin, Rosicky, Walcott, van Persie, Arshavin.
Subs: Fabianski, Oxlade-Chamberlain, Chamakh, Lansbury, Miquel, Ozyakup, Sunu.

No need to pile on Arsenal's misery here. For the first time that I remember, I don't own a single Arsenal player, and no one even registers on the shortlist. Van Persie took his goal well to remind us of his talent but the surrounding cast compared to Rooney and Aguero is just not there. We can reassess this team when Gervinho and Song return from suspension and (hopefully) new signings arrive.

Aston Villa
Given, Herd, Dunne, Collins, Warnock, N'Zogbia, Petrov, Delph, Agbonlahor, Heskey, Bent.
Subs: Guzan, Ireland, Albrighton, Delfouneso, Makoun, Clark, Bannan.

Luke Young has surprisingly moved to QPR leaving a hole to be filled at right back. If it is to be Herd (who played very well against Wolves) then this case is closed as he classified as a midfielder, but the situation needs to be monitored as if a new signing arrives or Beye (4.5m) gets significant minutes, it could possibly represent cheap access to a side with two clean sheets to their name already.

Offensively this looks like a settled lineup and McLeigh appears to be happy to stick with the 3-man front line. Despite the attacking look on paper, the goals haven't exactly been pouring in with two blanks accompanying the decent display against Blackburn. The concern is what this means for N'Zogbia who is forced to play much deeper and much narrower than he did at Wigan and chances from open play are going to therefore be limited. You'd be surprised if he didn't grab a couple of goals at some point this year but there is little justification for paying 1.0m more than the likes of Brunt or Jarvis, making him hard to justify owning.

Robinson, Salgado, Samba, Givet, Olsson, Emerton, Nzonzi, Dunn, Pedersen, Hoilett, Roberts.
Subs: Bunn, Formica, Petrovic, Rochina, Goodwillie, Hanley, Lowe.

Not too much to say here other than that David Dunn will likely resume penalty duties on his return. Blackburn have very little to offer from a fantasy perspective.

Jaaskelainen, Steinsson, Cahill, Knight, Robinson, Eagles, Reo-Coker, Muamba, Petrov, K Davies, Klasnic.
Subs: Bogdan, Sanli, M Davies, Blake, Pratley, Wheater, Riley.

It looks like Cahill will be staying where he is, so if anyone is holding him on the expectation of him playing for a stronger team in the future, I'd give that up now. Like last year this Bolton side cannot be relied upon away from home or against the top sides, thus limiting your starting opportunities to around half of their games played. This is fine as a rotation option but Cahill (at his inflated price) needs to be played every week, hence why I am not on board this season, despite his GW1 heroics.

Eagles and Muamba also drew plenty of attention after GW1 and while I was quick to dismiss Muamba (who actually took a knock here) I thought Eagles might be able to provide some value this year. It's harsh to judge a players playing against a good Liverpool side at Anfield but Eagles looked lost for long stretches of the game and the few set pieces he took charge of were not dispatched with particular quality. The arrival of Tuncay and the way Mark Davies has played off the bench (albeit in a central role) suggests that there is room for Coyle to reshuffle and there is no guarantee that either of these plays will be in the side for a sustained period of time. The price tags still look appealing but start to consider replacements now or be left flat footed in the future.

Hilario, Bosingwa, Ivanovic, Terry, Cole, Ramires, Mikel, Lampard, Drogba, Torres, Malouda.
Subs: Turnbull, Mata, Lukaku, Ferreira, McEachran, Alex, Anelka.

The Bosingwa bandwagon is sure to pick up speed this week after his excellent goal, but there is still remains a nagging concern that Ivanovic might steal minutes at right back when Luiz is fit. At 5.5m you could probably afford to have him sit every 4 weeks or so but if that time share is more like 50:50 then both will be tough to own this year.

Drogba (who suffered a mild concussion), Torres and Lampard all looked quiet once again and their lofty price tags make them poor choices at the moment. Drogba's injury might help Torres in the long term as he could get a few games as the lone frontman in a 4-5-1, but still, he looked timid and lacked confidence yesterday and I wouldn't go anywhere near him over the likes of Rooney and Aguero at the moment.

Mata looked sharp when coming on and took his goal a calmness that was lacking from many of his teammates. You would think that when the dust settles he emerges with a starting place in this side and at 9.5m he looks like he could represent the best value in Chelsea's front-six. Unless you're forced into it, it might be a week or two too early to make that move and GW6 (Swa H) looks like a good target date to make such a move.

Howard, Neville, Jagielka, Distin, Baines, Osman, Heitinga, Arteta, Barkley, Fellaini, Anichebe.
Subs: Mucha, Hibbert, Bilyaletdinov, Beckford, Cahill, Vellios, Baxter.

Any previous reader knows I'm not a big proponent of the high priced defender, and while I wouldn't take credit for knowing that Baines will lose penalty duties, it is another reason why it's so dangerous to pay a price which has already accounted for a high offensive points total. Baines may well take the next penalty but everyone he misses (or doesn't take) and with every game that goes by, the chance of him repaying that massive price tag diminishes, yet you feel compelled to hold him on the basis that he must be 'due' for a change soon. If you like him, you'll probably keep him but this should serve as another warning sign.

Schwarzer, Briggs, Hughes, Hangeland, Baird, Dempsey, Murphy, Sidwell, Duff, Dembele, Kasami.
Subs: Etheridge, Kelly, Johnson, Senderos, Gecov, Etuhu, Frei.

The points of interest here are more who didn't play than who did. Johnson and Senderos were developing into good looking squad players but were both benched here. Whether not this was a simple case of a one week rest (Fulham's season has been going on since early July remember) or a more concerted effort to rotate the squad remains to be seen. In both cases I think the low price tags mean they remain ownable in the short term but obviously if the rotation persists then new options needs to be found.

Reina, Kelly, Carragher, Agger, Jose Enrique, Adam, Lucas, Kuyt, Henderson, Downing, Suarez.
Subs: Doni, Carroll, Maxi, Spearing, Shelvey, Skrtel, Robinson.

Adam continues to play a fairly deep role but his set pieces and Liverpool's deliberate build up play allow him to stay involved and he has far exceeded my early season expectations. I'm still a bit cautious regarding owning Adam given the uncertainty of how the team fits together when Gerrard returns but on form to date one would suggest that it would be Henderson on the right or Kuyt/Caroll up top who will miss out. With so many premium options flattering to deceive, Adam deserves our attention once again.

Carroll's price tag (0.5m more than Suarez) made him unattractive initially but his early form and this benching make him even less desirable. Though Suarez is owned by large numbers of other managers, he is simply a better option for the short-medium term.

Man City
Hart, Zabaleta, Kompany, Lescott, Clichy, Toure Yaya, Barry, Silva, Nasri, Aguero, Dzeko.
Subs: Pantilimon, Richards, Milner, Johnson, Savic, Tevez, Balotelli.

We knew Zabaleta was going to get a game at some point and as it turned out this was probably a decent game for it to happen. I think the defensive rankings for City remain as you were with Kompany being the safest to play every week and Richards carrying the higher upside but more rotation risk (I think this rotation risk remains lower than Clichy though). The defense looks slacker than last year with Mancini loosening the reigns a little bit so far. While the unit remains well priced at 6.0m I am less high on them than I was to open the season and it seems likely you will better served with two attacking rather than two defensive options from this side.

As expected Yaya Toure looked much less offensively dangerous with the arrival of Nasri and his fantasy returns have taken a bit hit. Nasri's inclusion also took some of the attention away from Silva, though while that led to reduced points this week, over the course of season I would imagine there are plenty of points to go round for both, who look fairly locked in to get significant minutes each.

Up front, Aguero looks like as good of an option as anyone not named Rooney and that pair look to have distanced themselves from everyone else in the early running. Dzeko is one of the hardest players to forecast right now as the shadow of Tevez looms large but remained unused again today. If he's playing he is clearly outstanding value at 8.5m and with Balotelli apparently out of the picture completely, at worst Dzeko looks good for a couple more weeks of good returns. If you're happy with the rest of your squad his ceiling may be worth chasing, even if you end up having to cut him again in a few weeks.

Man Utd
De Gea, Smalling, Jones, Evans, Evra, Nani, Cleverley, Anderson, Young, Rooney, Welbeck.
Subs: Lindegaard, Ferdinand, Berbatov, Giggs, Park, Hernandez, Fabio Da Silva.

Though Nani had his best fantasy game of the season, he was once again overshadowed by a phenomenal performance from Ashley Young. We've discussed this several times before but it bears reiterating that Young is the better option in this side for now, and we can review this when Valencia returns to full fitness and some rotation will inevitably happen.

We've also said before that Rooney is the best option around at the moment and this game only underlined that fact. Penalty duty, set pieces and all around play in a great side all point to an elite option and his current lack of competition makes him almost impossible to drop.

Rio Ferdinand was almost fit enough to start this one and should return next week. That should push Evans (who didn't look too convincing) out of the side and end his run as a great budget option.

Krul, Simpson, S Taylor, Coloccini, R Taylor, Lovenkrands, Cabaye, Tiote, Gutierrez, Obertan, Best.
Subs: Soderberg, Smith, Ba, Marveaux, Sa Ameobi, Vuckic, Ferguson.

Not too much to add here. The defence remains a decent option though will always be unreliable as they proved this week with a decent fixture. The front options are fairly uninspiring.

Ruddy, Barnett, De Laet, Whitbread, Naughton, Hoolahan, B Johnson, Crofts, Tierney, Holt, C Martin.
Subs: Rudd, R Martin, Morison, Jackson, Pilkington, Fox, Bennett.

From what I have seen, no one is emerging from this team as a legitimate option right now and while Holt (who received some pre-season attention) took his goal well, I haven't seen enough to suggest he is anything more than a wild punt at the moment.

Kenny, Perone, Gabbidon, Hall, Connolly, Taarabt, Faurlin, Derry, Buzsaky, Smith, Agyemang.
Subs: Murphy, Bothroyd, Helguson, Ephraim, Hewitt, Andrade, Harriman.

I haven't seen their game this week, so I'll hopefully have more to add after reviewing the chalkboards/stats but until then, there are few options here to excite us so far.

Begovic, Huth, Shawcross, Woodgate, Wilson, Pennant, Whelan, Whitehead, Etherington, Walters, Jones.
Subs: Sorensen, Collins, Soares, Pugh, Upson, Wilkinson, Shotton.

The team looks settled now and the defence is once again emerging as a legitimate budget option. Woodgate (4.5m), while fit, is an excellent cheap link and should be strongly considered for anyone needing to move Evans this week.

Mignolet, Bardsley, Brown, O'Shea, Ferdinand, Cattermole, Larsson, Gardner, Colback, Gyan, Sessegnon.
Subs: Westwood, Wickham, Richardson, Vaughan, Ji, Bramble, Elmohamady.

Another lineup that looks fairly settled though isn't providing too much excitement. I thought Gyan caught the eye as their most promising attacking option and if there's no way you can stretch to the 9.5m range, he deserves some consideration. The fixtures look tough though so temper your expectations if you do make that move.

Vorm, Williams, Taylor, Caulker, Rangel, Britton, Sinclair, Dyer, Routledge, Graham, Agustien.
Subs: Moreira, Tate, Dobbie, Lita, Moore, Allen, Gower.

Swansea had a very good defence last year and the addition of Vorm has allowed that to transfer to the Premier League. The best part about Vorm (and Al Habsi below) is that while clean sheets won't always be there, his save totals should prevent too many 2 point performances and at 4.0m he looks like the perfect half to a rotation combo.

Friedel, Corluka, Dawson, Kaboul, Assou-Ekotto, Lennon, Modric, Kranjcar, Bale, Van der Vaart, Crouch.
Subs: Cudicini, Huddlestone, Pavlyuchenko, Defoe, Bassong, Walker, Livermore.

Defoe hasn't started the season too well and this game illustrates that he isn't even certain to start anymore. He doesn't even register on my shortlist right now. Van der Vaart and Bale didn't have the best of times but considering the opposition it's too early to right this talented pair off yet. Spurs' fixtures look better from here on and given the lack of attention they're receiving, there could be some differential options here.

West Brom
Foster, Reid, Tamas, Olsson, Shorey, Brunt, Scharner, Mulumbu, Morrison, Tchoyi, Long.
Subs: Fulop, Thomas, Dorrans, Dawson, Jones, Cox, Fortune.

I didn't see West Brom's game this week so I've not too much to add here. The players with fantasy potential (Brunt and Long) continue to play while Odemwingie is conspicuous in his absence. The teams deserves monitoring but clearly isn't firing on all cylinders just yet.

Al Habsi, Boyce, Caldwell, Lopez, Figueroa, Gomez, Diame, Watson, Moses, Rodallega, Di Santo.
Subs: Kirkland, McCarthy, Thomas, McArthur, Sammon, Jones, Stam.

Like Vorm, Al Habsi's true value lies in the fact that he can deliver clean sheets and also good save totals. A very good options in a rotation strategy.

Hennessey, Stearman, Berra, Johnson, Ward, Jarvis, Henry, O'Hara, Hunt, Fletcher, Doyle.
Subs: De Vries, Elokobi, Kightly, Vokes, Hammill, Milijas, Foley.

Like West Brom, I like the talent in this team but you have to accept that it won't click every week. The best fantasy options look to be fairly engrained in the side and it is especially encouraging to see McCarthy stick with the two men up top, even when away from home. The game didn't give us any clues as to which of these options will be the best for the remainder of the year, though it did suggest they will at least all get their chance to prove their worth.

Thursday, August 25, 2011

Gameweek 3 Preview

Two weeks is a long time in both fantasy and reality. If it goes well you could being talked about as title contenders with the world at your feet (Mancini) but if things go wrong, even the strongest of resumes will be ignored (Wenger). Hopefully you find yourself in the former camp, but if not, don't panic, decide who needs to go then start trying to rack up some points, starting with this week.

Clean Sheet Rankings
It's always tricky at this point in the season to determine how much attention to pay to prior year and how much to focus on the current season. I have avoided the tough decision by simply adding/averaging the data to give hybrid values. The GPG conceded will therefore be over the 19 home/away games from last year plus this year's single game while clean sheets/ failed to score numbers are simply added together:
[Defensive Pick vs Opponent - goals per game conceded (clean sheets kept - opponent failed to score)]
  1. Liverpool vs Bolton - 0.93 GP (9 - 7)
  2. Aston Villa vs Wolves - 0.95 (7 - 7)
  3. Chelsea vs Noriwch - 0.97 (11 - n/a)
  4. Blackburn vs Everton - 0.98 (7 - 7)
  5. Swansea vs Sunderland - 1.00 (n/a - 8)
  6. Newcastle vs Fulham - 1.15 (5 - 7)
  7. Everton @ Blackburn - 1.15 (3 - 1)
  8. QPR @ Wigan Athletic - 1.20 (n/a - 2)
  9. West Brom vs Stoke - 1.20 (2 - 10)
  10. Tottenham vs Man City - 1.23 (5 - 6)
  11. Man Utd vs Arsenal - 1.28 (11 - 3)
  12. Wigan vs QPR - 1.36 (3 - n/a)
  13. Man City @ Tottenham - 1.39 (6 - 4)
  14. Sunderland @ Swansea - 1.44 (6 - n/a)
  15. Stoke @ West Brom - 1.55 (3 - 4)
  16. Fulham @ Newcastle - 1.58 (5 - 4)
  17. Wolves @ Aston Villa - 1.65 (2 - 4)
  18. Bolton @ Liverpool - 1.75 (1 - 2)
  19. Arsenal @ Man Utd - 2.00 (5 - 0)
  20. Norwich @ Chelsea - 2.10 (n/a - 2)
Attacking Rankings

Chelsea (2.10 GPG) - having once given us two of the most reliable players of the past few years in Drogba and Lampard, this team is suddenly devoid of guaranteed fantasy talent. If you own Lampard or Torres then I guess if you don't captain them this week you never will but both have a huge risk of coming in with a two pointer, not really felt in prior years. Norwich have done okay against two weaker sides this year but this will be a good game for both teams to see where they are. I'd prefer not to be starting players in a game we are designating a test but the appeal is there and I'd certainly be tempted by Torres, Lampard or even Malouda this week.

Man Utd (2.00 GPG) - it's surprising to see United this high given their strong opponent, considering that this data is based largely on prior year. But the reality is that Arsenal did not travel too well last year (1.47 GPG, 8th) while United were simply sensational at home (2.58 GPG, 1st). With Koscielny a doubt for this one, the defense looks as thin as the midfield unit and you would imagine United will be able to breach the back line a couple of times. It's really tough to look past Rooney here (indeed I'm fairly sure that's the direction I will go in) but I think Young has just about shown enough to play his way into contention though I would suggest than Sagna is a step or two above either of the full backs he has faced so far this year.

Liverpool (1.75 GPG) - the biggest question is whether Suarez plays, a question not helped by his surprising inclusion in tonight's Carling Cup game. With his rest at the weekend, one assumed he was still a little fatigued from the Copa America, so quite why he was needed against Exeter is beyond me. The risk of Suarez sitting again would not deter me from ownership but it might from handing him the armband. The concern I have is that if he sits, he still probably comes on for 20 minutes or so at the end, thus preventing your VC from stepping up. If we get confirmation that he will play (or is nearly certain to) by Friday then he jumps up the rankings and could be argued to be alongside anyone this week.

Elsewhere, your options are probably limited to Suarez's up front partner, who also scored in the Cup midweek. I've been a little higher on how Carroll has played so far this year than most and if I owned neither Liverpool strikers but wanted Suarez long term, I might well buy Carroll for this week, moving to Suarez a bit later when his place in the side in safer. If Suarez plays, the previous paragraph looks foolish but fantasy is about taking calculated risks so be sure you like your odds before giving Suarez the 'C' this week.

Man City (1.39 GPG) - not the best fixture for the league's hot team and if you held off buying Aguero last week, I'd probably do so again this week, with Rooney, Carroll and Bent all enjoying better fixtures and apparent guaranteed minutes. On the off chance you don't own any other good options this week, I'd take Silva as the man to back from this unit given his ability to contribute fantasy points from a number of categories but I'm lower on him than I was in the previous couple of GWs.

Elsewhere, Darren Bent (Villa 1.65 GPG) has a great fixture once again and looks to be the best pick from a non big-five team. Bent scored just 2 of his 7 goals for Villa last season at home but he was able to add to that last week. Wolves haven't been great away from home but they did win the corresponding fixture 1-0 last season and have looked solid at the back so far this year.

Reader Questions
Alex - Wanted your opinion on who's a better buy at this point: Klasnic, Fletcher, or Long.
On pure ability I would probably take Fletcher but his fixtures look pretty tough in the medium term which is a big factor for a mid-range forward. Klasnic's fixtures are better than they look on paper as they get the majority of the tough teams at home, but the upside is still significantly below West Brom's. If you are looking for an every week starter, it's tough to pick between any of this trio, but if you would consider rotating them, then Long has the most upside fixtures in which you would definitely start him.

Bobby Reds - I still cant decide between Brunt & Jarvis for the long term though
Another close call and again I would defer to the disparity in strength of schedules to make my decision. With Odemwingie ready to return to the West Brom lineup, their team looks fairly good and Brunt should once again be at the heart of a lot of their success. That 6.0m range is starting to look fairly attractive in midfield.

JS Shadow - are you going to post when would be the appropriate time to use the wild-card and how I should use it.
I will probably look at this in and around GW8. Normally teams may wish to use their WC earlier but the free transfers in GW2 allowed a lot to get rid of all their preseason mistakes. In the meantime, some thoughts on the issue from last season can be found here.

Shedboy - which is better? Torres/Aguero and Odemwingie - Rooney and Klasnic
Rooney is the safest elite player right now, though Aguero is probably right on his shoulder when the fixtures are equal. For this week Aguero doesn't look like a great play and Odemwingie may not be fully fit so the latter pairing looks to be preferable. Long term, you might think about moving Rooney to Aguero to free up some cash which brings Zamora, K Davies, Rodallega and Saha, among others into play. If you can spare an extra 1.0m, you could also try a combo of Suarez/Bent/Hernandez/Carroll which would be very interesting.

Amtosh - Would you turf Torres for Aguero, considering Torres has Norwich at home and aguero has spurs away?
I wouldn't be buying Torres at this point but if you own him you presumably did so with the intention of holding him for at least GW2 and GW3 and that still appears to be the logical move here. Aguero has the talent to score against any opponent but Chelsea are usually very good at home and the Norwich fixture is just too good to pass on. Using another transfer only locks in Torres' lack of production so far.

Ryan - I've just been looking at my attacking options for my team and realised i have no attacking contributions from liverpool, arsenal, chelsea or tottenham. Do you think this is a wise choice?
It's hard to say in a vacuum but I wouldn't say it's necessarily an issue. I personally don't hold anyone from these teams either due to a combination of tough fixtures and, honestly, a lack of fantasy options from the London clubs. Chelsea's team is totally up in the air with the arrival of Lukaku and Mata and the ongoing Drogba/Torres dilemma while Arsenal would appear to have taken several steps back from last year. I like Bale for Spurs but I'd like to see Spurs show a bit more before I invest there.

D - I have Zamora, Alex, Moses, and O'Hara and I want to trade them all. What do you prefer between Boswinga/Cole and Odem/Long? 
I would probably bail on Alex at this point as you would imagine that Luiz retakes that starting spot this week. Zamora, Moses and O'Hara all have value so while you might consider looking elsewhere, you're in a good position in that with all three starting you have the ability to change your team over time. I'm still a bit uneasy about Bosingwa and you have to accept the risk of him being benched sometimes but he seems to have won favor over Ivanovic, and at worst it would seem to he will play against the weaker sides. For safety I'd take Luiz (provided he starts this week) but Bosingwa's upside is unparalleled. Right now I'd take Long as he's shown enough in his positioning and finishing ability to suggest he can succeed in the Premier League while Odemwingie is still not fit and struggled to get in the game last week when coming on. Perhaps do Alex to Bosingwa now and then wait to see if Odemwingie can show some form before choosing your West Brom forward.

@fplsavant - Seeing as Richards just got an assist for Silva's goal, do you think its time I consider him, or is Zabaleta a threat?
He seems to have moved into the safe category now, and while he may suffer the odd rest now again for Zabaleta you'd guess that (a) this happens less than with Clichy and Kolarov and (b) occurs when City face tougher sides (given Zabaleta's more conservative approach). He still isn't as safe as Kompany but his upside could very well push him into position as the best City defender.

@thatbensonguy - Can I why has Arsene Wenger waited til now to make enquiries? He was always losing Nas and Fab! He could have checked the market!
Not fantasy related but I'm a sucker for all things Wenger. I always defend Arsene and the criticism he receives is ridiculous but I agree with you on this one. I will never go along with the notion that he should be chasing Downing for 20m but he needs to at least be looking at someone. It will be a huge shame if Arsenal fail to strengthen in this window, and it would be indefensible for Wenger. I still trust he knows what he's doing but he's currently looking as unprepared as he ever has.

@JennineBorrow93 and @RonnieLeclere790 - as they're the same price, over the next few weeks do you go for Nani or Young?
Right now it has to be Young. I'm not ready to give up on Nani for the season yet and I'd like to see how the side gels with Hernandez set to return but the team is clearly focusing play down Young's side of the field and his set piece duties put him over the edge.

@JamesGall93 - I don't know who to pick in goal this week, Vorm or Al-Habsi?
It's a close call and the stats suggest Al-Habsi has the edge. Swansea look like they may have a better defensive unit than Wigan but QPR appear to be a weaker side than Sunderland. The stats suggest Al-Habsi has the better shot at a clean sheet and both keepers bring good save totals to help give some value if they do concede. If you're playing the rotation strategy you need to trust the system so I'd take Al-Habsi.

@NikosEllinas - Think Odemwingie will start? Is it time for K. Davies to Odem, have 1 min in the bank.
I wouldn't be thrilled about bringing Odemwingie in until he proves his fitness but with Davies facing such a tough fixture your hand is forced a bit. If we can get any kind of assurance he will start this week I would probably role the dice with the West Brom forward but without further notice take the points with Davies.

Wednesday, August 24, 2011

The Fantasy Fallacy

The Gambler's fallacy is a well documented theory which, loosely put, sets out that just because something has happened 20 times in row, the chance of it happening again is not diminished. In gambling we limit this to non-dependent events like coin tosses or roulette spins, but there is an application to fantasy scoring too.

Fantasy scoring is not totally non-dependant as players get in form and teams gel as a unit, but there is certainly a case for not waiting through 8 barren gameweeks on the basis that a player is 'due' for a big game to turn it all around.

With this in mind, I am introducing a new list of players who I have no problem cutting loose at this point in the season. I am not suggesting that they have no value for the year, but based on the immediate future, better options can be found and holding them may cost you weeks of wasted cash. Following this strategy may cause you to miss a big week from a player as any downturn from a top player is temporary, but on average, I would prefer to bail out and reinvest a week too late, rather than sit through week and after week of poor production.

You will see three lists below:
  • Cut 'em loose - too many factors weigh against these players and they should be sold immediately unless another player on your team is in an even worse predicament.
  • On the edge - probably should be sold but it isn't a huge priority due to one of a number of factors (their price tag isn't too high, their slump has only just started, concrete evidence of a turnaround can be seen)
  • Keeping tabs - the evidence is stacking up but I'm not ready to bail as we can at predict, with some evidence, an improvement in the future.
Cut 'em loose
Didier Drogba - it's unconscionable to hold a 10.0m+ player who is not even in a team's starting lineup. 124k managers still do though and with players like Bent, Suarez and the returning Hernandez available for less money, there is no excuse. Torres doesn't look great either and Drogba might return sooner rather than later, but at such a steep price, you simply cannot afford to wait, even though the fixture looks great this week. Just be thankful you didn't draft him in the 3rd round.

Dirk Kuyt - he got the start at the weekend but you'd think that's more a case of Suarez having the occasional rest rather than a blueprint for the season. Moving away from the 4-3-3 has eliminated Kuyt's starting spot, and when he does play it will likely be in a deeper role or at the end of a game. Like Drogba, the price means you simply cannot wait to see if Kuyt can win his place back.

Jon Arne Riise - if Lampard's owners are stuck in 2008, Riise's may be stuck in 2002. We all uttered some form of 'what if' when the signing was announced but it's clear after just a couple of games (and Euro appearances) that Riise is not the same attacking player playing in the same system. This is by no means an attack on his play, but he simply isn't going to score 6 goals a season as he did in his Liverpool days. He may well get a couple of goals and the odd assist but for 1.5m for than Senderos, there is no way you can justify the extra cost. Turn off your Maroon 5 CD and cut the chord with 2002 (in fact, even if you want to keep Riise, just turn of your Maroon 5 CD).

Demba Ba - registered as a safe pick in the pre-season but looked awful in GW1 and was benched in GW2. Over the course of the season he certainly has the ability to bounce back, but you cannot afford to have players on the bench at this point and other options are available in his price bracket who have found some early season form. Let's not write him off for the season but you need to elsewhere for now.

On the edge
Robin Van Persie - this one is as much about Van Persie's teammates as his own play. He's done okay in fantasy points without Fabregas and Nasri in the past, but that has tended to be over shortened periods where other players can temporarily take up the slack. As I write this, Van Persie is cooling off after scoring against Udinese and by all accounts was busy throughout. His cost means that he will hold onto his status on the edge, but there are at least signs that Van Persie might be ready to get his (and Arsenal's) Premier League campaign. Whether or not you hold onto him depends on your appetite for risk (Arsenal's fixtures look marginally better than Man Utd's but Rooney remains the safer bet for 12.0m).

Frank Lampard - he's playing a deeper role, his team don't look very convincing and he isn't getting a sniff at bonus points so far under the new system to bail him out as in the past. 70k managers are somehow affording to carry his corpse, presumably in a 'no man left behind' effort for his previous campaigns from '08-'10. Admirable as their loyalty may be, if Lampard doesn't produce a big game against Norwich this week, it's time to move on.

Fernando Torres - last week's transfer activity was somewhat strange as there was a lot of buzz generated around Torres, after the GW1 0-0 with Stoke. Yes, he was in the team at Drogba'sLampard above, someone in this Chelsea team is going to be valuable at some point and if you own them, you keep them for this week but the leash needs to be getting shorter by the day.

Louis Saha - the knock on Saha has always been his fitness. He appeared to be fit and ready to go in GW1, only for the game to postponed, but then had to sit out the majority of GW2. Again he could only feature on the bench for tonight's Carling Cup game, and the old story is starting to become repetitive. Everton face a tricky trip to Blackburn this week so cutting the cord now might be the best idea but if you still have faith in the Frenchman, you can't give him more than a couple of games to prove himself.

Keeping tabs
Nani - I was higher than most on Nani in the preseason and while I maintain that his playing time will continue to be fairly consistent throughout the season (not starting every game but getting decent minutes) his fantasy ceiling appears to be declining. Against Spurs, the play seemed to flow down Ashley Young's left side for long periods of the game with Nani forced to slide inside to even get a touch of the ball. With Young monopolizing set pieces too, Nani is losing value by the day and if I was choosing one to own I would side with Young right now. Is it worth a transfer though? I am giving it another couple of weeks given Nani's proven goalscoring record and I want to see what, if anything, changes when Hernandez gets into the side.

Adel Taarabt - Last week's 2 bonus points failed to appease my concern for Taarabt, who did little based on the eyeball test or chalkboard review. He looks well place to be the main beneficiary in this department whenever QPR win, but given that this could happen 10 times or less, the ceiling seems to be moderate and without significant goal totals I don't like him at 6.5m. At this status shows I am not ready to give up just yet but don't hold on too long as players like Jarvis, Brunt or Cleverly go unnoticed.

The full week preview will be posted shortly, so post your questions below or @plfantasy and they will be added to the new reader question section.

Monday, August 22, 2011

Change we can't believe in

As the saying goes, the more things change, the more they stay the same. We speculated in the pre-season about what differences the new bonus system might produce, given that it was now to be based on a scientific and objective formula. The EA Sports index is based on six categories of contribution, which combine to give a player's overall contribution to a game. The approximate weighting appears to be as below:

  • Scoring a goal: 90%
  • Winning the game: 10%
  • Other: 1%*
*Rounding difference of +/- 1% is possible

Only two of the top 12 scorers in GW1 failed to find the net (Kompany and Yaya Toure) while all found themselves on the winning team. In GW2 an incredible 13 of the top 20 scorers all benefited from having a goal to their name while the top 17 scorers all won their respective games. 

The top 20 scorers to date gives us 6 players who are yet to find the net - Richards, S Taylor, Adam, Yaya Toure, Bosingwa and Kompany. Four of these names are already highly prized fantasy assets anyway but Taylor and Adam at least bear mentioning:

S Taylor - these strong displays, along with his goal threat from set pieces go a long way to suggesting he may be worth the extra 0.5m over Danny Simpson. This theory is further cemented by the chance that Simpson could face a rotation threat from R Taylor if a full time left back is brought in over the next couple of weeks.

Adam - Adam was obviously great last year but a large portion of his value derived from his ability to attract bonus points in a weak Blackpool side. If he can continue to be rewarded under the new system (despite not scoring or adding an assist in GW2) then he may have value again this year.

In short, there is very little more to say on bonus points and it appears they will simply follow the goals and clean sheets you are already chasing. Perhaps we give a slight nod to the teams who are more likely to win, but given the extra cost you will have to pay to acquire such players, even that looks like a dead end.

Until something changes with the data we are seeing, I won't be touching on this area again for a long time.

[I'm not sure it's obvious reading this but the percentages were a joke, sort of]

SimPrem - Meet the Teams: Oglethorpe FC

I've touched on the new SimPrem game before, and now we're a couple of weeks into the season, I thought it was time to meet the teams and see how the draft went. We start with one of the co-founders, John Wright, owner and manager of Oglethorpe FC.

What was your strategy going into the draft?
As one of the two founding members of the SFP, I felt like I had a decent feel for what I would need going into the draft and perhaps a leg up on the opposition. Last year, I favored players who would receive more guaranteed playing time, only to find my team lacking quality at times. If not for a couple of lucky hits like Robin van Persie and Ben Foster, I probably wouldn't have won the league.

This year, I came into the draft favoring quality over quantity. The biggest constraint in the current SFP system is that you are limited to 90 minutes of action per week at 11 team places, just like real football. With 25 players to fill those spots, I knew that each week, I would have some mix of "matchup" type players who would fill minutes and "quality" type players on whom I could rely for goal and assist production on a per-minute basis. Of course, the best players mix both quantity and quality, so Wayne Rooney was at the top of my draft board as a player who is a near-lock to be in the top five for goals scored while playing a lot of minutes for a top team. Having the eighth pick, I was unlikely to get him, but I settled for another top striker in Andy Carroll.

The next biggest constraint is depth by position. I wanted a balance of players at each position, since the best formation bonus comes from a balanced formation like a 4–4–2 or 4–3–3. The top forwards and attacking midfielders are inevitably the class of the league, but forwards in particular are very difficult to replace with any quality, so they received the biggest positional adjustment. Defenders from the big six clubs received a boost from me, since they're less vulnerable to matchup problems. One other difference worth noting in my strategy compared to others is that I gave quality keepers considerably less weight than most others in the draft. The main reason for this is the unpredictability that I perceive in predicting saves, and thus, predicting who is actually going to be a quality fantasy keeper.

A side note about keepers: without better data from a source like Opta, there's not a good way to construct a scoring system to favor the actual best keepers, since the worst will get plenty of "cheap" saves, and even a poor keeper on a good team will allow relatively few goals with a good back line in front of him. In our system, a keeper is going to score a lot of defence points, more than any regular defender, but the difference between the top keeper and the worst is fairly small for me, given the unpredictable nature of the position.

Did you have a favorite pick?
My second round pick (9th overall) is where I probably started to divert from the conventional wisdom, taking a relative part-timer in Rafael Van der Vaart, who is a prolific fantasy scorer when he plays. The same could be said of Theo Walcott (Rd 3), Rafael (Rd 7), Salomon Kalou (Rd 9) and Steve Fletcher (Rd 11). So, to some extent, I'm relying on their minutes overlapping enough for me not to have a problem filling out a team from week to week. On the other hand, I felt like these were great values in the draft.

Some of my favorite picks from other teams were Didier Drogba with the 19th overall pick and Nani with the 16th pick, among the earlier selections. Later in the draft, I liked Clint Dempsey (Rd 7, great attacking midfielder for our format), Adam Johnson (Rd 13, quality in short bursts), and Benoit Assou-Ekotto (Big Six defender in Rd 19). Johnson is probably the player I wanted most at the time he was taken.

Any reaches?
My biggest reach was probably Emmanuel Adebayor in Round 18, given the uncertainty that he will play in the Premier League at all.

Lessons so far
I've learned over the years playing fantasy games not to overreact to one week's results, so you won't find me overrating Chris Eagles or the Bolton defence based on one match against a newly promoted club. The biggest lesson I think you can learn from the first week is which players might be first choice at their positions in cases where playing time was previously uncertain.

It was revealing to learn that Steve Fletcher, in keeping with his late-season form from 2010/11, was chosen to start and play the full 90' in the same team with Kevin Doyle, especially in the event that Doyle doesn't move to Arsenal as rumored.

All in all, I am happy with my team. Now it's all about setting the right tactics for each week and bringing in the right players to fill my smaller needs.

Oglethorpe opened the season with 0-0 draw in week one followed by a 3-0 loss to a good looking Cat & Moose FC, owned by the excellent team over at Never Captain Nicky Butt.

Sunday, August 21, 2011

Gameweek 2 Lineup Lessons

Szczesny, Koscielny, Vermaelen, Jenkinson, Sagna, Frimpong, Ramsey, Nasri, Arshavin, Walcott, van Persie. Subs: Fabianski, Oxlade-Chamberlain, Chamakh, Miyaichi, Lansbury, Miquel, Bendtner.

Not too much to talk about here as Wenger basically picked all the first team players who were available to him. Anyone who reads this blog knows how much I like Wenger and respect Arsenal but it's becoming increasingly hard to get enthusiastic about anyone in this team from a fantasy perspective. For all the talk this week of Torres v Aguero v Rooney, Van Persie was conspicuous in his absence. I have no doubt he will have success at some point this year but I would bail for now and come back when this team can string some consistent performances together.

Aston Villa
Given, Young, Dunne, Collins, Warnock, N'Zogbia, Delph, Petrov, Agbonlahor, Bent, Heskey. Subs: Guzan, Ireland, Albrighton, Delfouneso, Makoun, Clark, Bannan.

On this evidence it would seem that McLeigh might stick with the 3-man front line which hurts N'Zogbia's value a bit. He was more involved than last week but he was mainly deployed on the right and hence forced to cut inside which severely limits his ability to play dangerous balls into the box. Bent didn't play great again but he got his goal, and that tends to be what we expect from Bent. He's never going to be a player who stands out on chalkboards or dictates a game but his conversion rate of chances to goals is one of the best in the league. I still rank him a step below the elite forwards but well worth a one or two week rental when the fixture looks right (as it did this week). Albrighton came on at half time but until further notice he should not be owned.

Robinson, Salgado, Nelsen, Hanley, Olsson, Hoilett, Dunn, Nzonzi, Pedersen, Goodwillie, Roberts. Subs: Bunn, Formica, Petrovic, Rochina, Blackman, Ribeiro, Lowe.

A chunk of managers jumped on Formica after last week's goal but this benching should scare the majority of those off again. The arrival of Petrovic further clowds this midfield situation, making no one particularly attractive in the short term. There is potential there though, so if a settled lineup emerges, someone among this group could be worth looking at.

Jaaskelainen, Steinsson, Cahill, Knight, Robinson, Eagles, Muamba, Reo-Coker, Petrov, Kevin Davies, Klasnic. Subs: Bogdan, Alonso, Mark Davies, Blake, Pratley, Wheater, Blakeman.

Shaping up to be one of the more settled lineups in the league with a clear distinction between the starters and the bench players. The other source of question is the central midfield but I don't really rate any of these as fantasy options anyway (tough Reo Coker has some use a 4.5m player). I liked Klasnic coming into the season and he's now got two in two. After week one I was a touch concerned by how he dropped off the front too often but at home he looked far more dangerous, taking his goal very well and generally looking quite lively. The fixtures would still need to fall right, but I can see a situation where at 6.0m he gets serious consideration this year.

Hilario, Bosingwa, Alex, Terry, Cole, Ramires, Mikel, Lampard, Anelka, Torres, Kalou. Subs: Turnbull, Ivanovic, Benayoun, Drogba, Malouda, Ferreira, McEachran.

Hilario proved he was the choice over Turnbull but it was all for nothing as West Brom took a surprise lead at the Bridge. Chelsea looked fairly poor all game and, like their London rivals, it is hard to see too many attractive options from this unit at the moment. For all the excitement this week, Torres regressed to the form of last year, looking utterly lost and failing to register a single shot all day. Anelka looks to be the brightest of this front three (with the benched Drogba) but none of them seem to be elite options.

One bright spot was the offensive play of Bosingwa who had over 100 touches and got into good positions throughout. His cross for the second game was awesome and at 5.5m he has the potential to be great value. I was worried that Ivanovic would take his place back this week, but at least for now it seems that Bosingwa is first choice. I wonder if this might be a platoon situation where Bosingwa plays against weaker teams while Bosingwa starts when facing elite opponents (the Young/Nani combo in GW5 will be a good test of this theory).

Howard, Neville, Jagielka, Distin, Baines, Barkley, Heitinga, Rodwell, Osman, Cahill, Beckford. Subs: Mucha, Hibbert, Saha, Arteta, Fellaini, Vellios, Anichebe.

Beckford over Saha was a puzzler, as the former Leeds forward has been linked with a move away from Goodison in the past few weeks. The 7.0m - 9.0m price bracket for forwards is crying our for someone to excel and Saha appears to be another name out of the running. There was some suggestion of Hibbert being a cheap link in this side but Neville took the right back duties despite Hibbert being available, and so that strategy looks to be unhelpful. It was reported that ArtetaCahil resumed his duties playing off a lone striker and while yesterday proved fruitless, he should be a fantasy factor once again in that role.

Schwarzer, Hughes, Hangeland, Senderos, Riise, Duff, Etuhu, Murphy, Kasami, Dempsey, Johnson. Subs: Etheridge, Kelly, Sidwell, Gecov, Frei, Briggs, Dembele.

Another teams whose lineup looks fairly settled with the core players appearing to be locked in. Baird missed out altogether which solidifies Senderos as a great value piece going forward. I liked Dempsey's deployment much more this week, though he still only registered a single shot on goal. He's another player caught between the elite and the bargains, and it looks increasingly hard to own him.

Reina, Kelly, Carragher, Agger, Jose Enrique, Lucas, Henderson, Adam, Downing, Kuyt, Carroll. Subs: Doni, Meireles, Suarez, Maxi, Spearing, Skrtel, Flanagan.

Not sure what to do with Suarez's benching here. I felt Kuyt would miss out on midfield in the pre-season and with Gerrard back in training, that situation will only get worse for him. The concern for fantasy owners is if he becomes a third striker and we get a situation where he plays once every 3 or four games up front. Given Suarez's price tag, he is still very ownable even if he does get rotated but it is a small blemish on his increasingly attractive resume.

Man City
Hart, Richards, Kompany, Lescott, Kolarov, Milner, Barry, Toure Yaya, Silva, Dzeko, Aguero. Subs: Pantilimon, Zabaleta, Clichy, Savic, Johnson, Tevez, Balotelli

Richards confirmed that he has slightly more value that Clichy, as Kolarov is more of a threat than Zabaleta. Expect Richards to receive his bench stint in the next few weeks, but over the course of a season he still looks favorite to record more starts. The front line looks more crowded with the return of Tevez, though Aguero would now seem to be locked in to at least receive 60 minutes (with the occasional rest). Your choice this week if you don't own him is to either bring him in and play him in an unfavorable fixture (at Spurs) or, wait a week but pay an extra 0.1m or 0.2m to get him. Silva confirmed my pre-season suggestion that he is the number one midfielder around (regardless of price) and should be owned by everyone, regardless of fixture, at least until we see what happens with Nasri.

Krul, Simpson, S Taylor, Coloccini, R Taylor, Tiote, Cabaye, Barton, Obertan, Gutierrez, Shola Ameobi. Subs: Harper, Williamson, Lovenkrands, Gosling, Ba, Best, Marveaux.

Ryan Taylor is doing his best to hold onto that left back slot, and if he does, he represents outstanding value for 4.0m. We won't be sure on that until August 31 so for now I would hold on Simpson or S Taylor. Ba was totally anonymous is GW1 and found a place on the bench in response. That front line looks fairly patchy and it's hard to see too much value there right now. Harper was available for this one and still found himself behind Krul, so I think we're safe to now assume the #1 jersey is Krul's.

Ruddy, Naughton, Barnett, De Laet, Tierney, Bennett, Crofts, Bradley Johnson, Pilkington, Holt, Chris Martin. Subs: Rudd, Russell Martin, Jackson, Surman, Hoolahan, Fox, Ayala.

Big changes here for Paul Lambert, bringing the value of most of the team into question. Goalscorer Hoolahan and Fox (who impressed me) were both benched giving the team just four outfielders who have played 180 minutes (Tierney, De Laet, Crofts and Holt). Pilkington, Bennett and Naughton, along with Ayala are all new arrivals for the Canaries so it would seem likely that some, or all, of these stick in the first team. It's safe to say so far that there's no Chaz Adam 2.0 so I might bail on this team for the immediate future until we can sure which budget players might deliver decent value.

Kenny, Orr, Hall, Gabbidon, Connolly, Buzsaky, Derry, Faurlin, Smith, Taarabt, Agyemang. Subs: Murphy, Bothroyd, Gorkss, Perone, Moen, Ephraim, Andrade.

Campbell and Borthroyd were dropped/benched here because of a virus so there isn't too much to read into that. Still not really convinced by Taarabt as a big-time fantasy producer, registering just one shot and taking up a fairly deep position on the left for large portions of the game. I would think he will have a nice run at some point, but at 6.5m you are paying a potential premium over players like Jarvis and Etherington who have enjoyed success in the Premier League before.

Begovic, Shawcross, Huth, Woodgate, Wilson, Pennant, Whelan, Whitehead, Etherington, Jones, Walters. Subs: Sorensen, Collins, Pugh, Upson, Wilkinson, Shotton, Arismendi.

Another team that looks fairly settled, suggesting that Woodgate will start most weeks (at least when healthy). Not much else to note here, so moving on.

Mignolet, Bardsley, Brown, Ferdinand, Richardson, Larsson, Cattermole, Colback, Elmohamady, Sessegnon, Gyan. Subs: Westwood, Gardner, Wickham, Vaughan, Ji, Bramble, Laing.

Sessegnon's positioning is encouraging but still not end product as yet. I'm a touch worried about this team scoring goals, having not created two many chances in 180 minutes. Gyan will surely get his goals at some point but how many Larsson and Sessegnon chip in with still looks uncertain. The fixtures don't look to be much help thus making this whole unit less attractive.

Vorm, Rangel, Caulker, Williams, Taylor, Agustien, Britton, Routledge, Dyer, Sinclair, Graham. Subs: Moreira, Tate, Dobbie, Lita, Moore, Allen, Gower.

Goals are going to be a problem but Vorm looks like a very good keeper and they might generate enough clean sheets to justify his ownership in a rotation system. I like his prospects given that his high save totals should bail out some of the games where goals are conceded. The lineup looks fairly settled but with no goals and not too many chances through 180 minutes it's tough to be overly high on the anticipated pair of Graham and Sinclair. Graham and Dyer's price make them ownable but I might move on from Sinclair until we see he can do it in the big leagues.

West Brom
Foster, Reid, Tamas, Olsson, Shorey, Scharner, Mulumbu, Brunt, Morrison, Long, Tchoyi. Subs: Fulop, Cech, Dorrans, Reyes, McAuley, Odemwingie, Cox.

A very settled team which should see Odemwingie slot in for probably Tchoyi next week. The Nigerian looked fairly sharp when coming on and should have equalised for the Baggies. Long's going to get a lot of attention with goals against United and Chelsea, and with Klasnic and Doyle, he forms a nice trio of potential fantasy targets in the 6.0m bracket.

Al Habsi, Boyce, Caldwell, Alcaraz, Figueroa, Diame, Watson, Gomez, McCarthy, Moses, Di Santo. Subs: Pollitt, Thomas, McArthur, Sammon, Jones, Rodallega, Stam.

Rodallega not starting is a big problem as his consistency and guaranteed minutes are a couple of his main selling points. Whether or nor this was linked to a poor performance last week or talk of a potential move away from the DW is unclear. Until this situation is resolved, opt for for of the aforementioned 6.0m guys or look to the ever reliable (and undroppable) Kevin Davies instead.

Hennessey, Stearman, Johnson, Berra, Ward, Jarvis, Henry, O'Hara, Hunt, Doyle, Fletcher. Subs: De Vries, Elokobi, Kightly, Ebanks-Blake, Hammill, Milijas, Foley.

A good settled side here, and I like this midfield a lot. Kightly and Milijas figure to take some minutes at some point but until then it's more a case of who you like more out of Jarvis, O'Hara and Hunt. Jarvis is obviously going to get the attention with a goal and a couple of assists already but I like the other two to contribute at some point too. If I needed a guy now I would probably just ride the hot hand with Jarvis.

Saturday, August 20, 2011

Defensive Fixture Combinations

I've posted at length before about the GK rotation strategy, with the basic idea being that good keepers with great fixtures tend to perform better (and score more points) than great keepers with average fixtures. To maximise the strategy you need two teams' fixtures to 'gel', that is, maximise the number of home fixtures and weak opponents and minimise the number of times both teams face a tough opponent.

Though I have always focused this strategy on keepers, there is no reason it cannot be used for defenders, and I'm sure many of you do it already. The availability this year of players from good mid-range teams for a budget price suggests that this strategy might be even more fruitful this year. With that in mind, I have expanded the GK-ranking combinations to include all teams which contain a regular starter available for 5.0m or less (thus excluding 6 teams - Man City, Man Utd, Arsenal, Liverpool, Everton and Chelsea). I know Liverpool do have a 5.0m starter in Jose Enrique but I figured you are realistically going to start him every week and including an elite side like Liverpool would potentially skew the data a bit.

The below shows the average number of goals per game predicted to be conceded, having taken the best fixture from the two teams each week. The period reflected is the next 10 gameweeks. I will update these pairings each week and you will be able to find the table in the 'Strength of Schedule' tab.

As with all early season data, it is largely backwards looking with only minor tweaks made for player transfers and manager movements. The promoted teams numbers are based on their Championship performances, adjusted by a historic ratio of how teams perform when promoted. Hence, don't put too much stock in a certain combo being 0.02 better than another, but rather note where a combo clearly doesn't appear to work.

As a basis for comparison, here are the average predicted goals conceded over the next 10gameweeks for those six excluded teams:
  • Man City 1.17
  • Chelsea 1.19
  • Arsenal 1.22
  • Man Utd 1.23
  • Liverpool 1.27
  • Everton 1.32
It isn't quite as simply as this, and there's something to be said for being able to plug in a Kompany or a Ashley Cole every week, but at least in theory, picking the best fixture from a Stoke/Villa combo should outperform playing a City defender every week (the difference between conceding less goals and keeping clean sheets is an argument for another day). As I say, it is full proof and shouldn't be used as a iron clad rule, but I hope it provides some guidance when filling out your squad players.

Thursday, August 18, 2011

Gameweek 2 Preview

The pre-season is the most hectic time for this blog, as setting a good initial lineup is a big - though not essential - step towards a good season. I had hoped to be able to take it easy for the first few weeks, assess some potential targets and start crunching numbers again after we've seen a few trends. Then all hell broke loose. There's no need to rehash the week's events but the end result is basically a press of the reset button with a few people enjoying an advantage of having a decent opening week (personally I was able to make the changes I wanted and just made a couple of picks that didn't work out).

In case you missed any of them during the week, I will shamelessly plug the previous pieces on week one lineup lessons, a bit of chalkboard analysis, contrarian views to the most targeted players and potentially cheap links in team defenses.

With that behind us, let's get onto this week. All the below rankings are for this week only so if you're still moving players around with your free transfers, refer to the long term strength of schedules rather than the below.

Clean Sheet Rankings

Rank Team Opp GPG CS Opp FTS
1 AR LI 0.97 9 5
1 MU TH 0.97 10 4
3 SW WA 0.97 -- 7
4 CH WB 1.03 11 9
5 EV QP 1.09 6 --
6 SU NU 1.11 7 7
7 AV BR 1.13 7 7
8 WW FU 1.29 3 6
9 BW MC 1.32 5 6
9 FU @WW 1.32 5 3
11 QP @EV 1.42 -- 2
12 WA @SW 1.43 4 --
13 MC @BW 1.45 6 4
13 NU @SU 1.45 5 4
15 SC @NC 1.62 3 --
16 LI @AR 1.66 5 4
17 NC SC 1.76 -- 10
18 BR @AV 1.82 1 4
19 TH @MU 2.00 3 0
20 WB @CH 2.11 0 2
As you will have worked out, goals per game (GPG), clean sheet (CS) and failed to score (FTS) data are based on prior year as one week of data for 2011/12 is obviously not enough to go on. As with last week, I urge you to exercise caution when it comes to the promoted teams' data as this is based on a convoluted historical method which may or not translate to reality (the rankings between the three teams should be useful though).

Based on last week's performance, QPR are currently overestimated in my stats and so Everton and the rest of the top-5 probably distance themselves a little from the rest of the pack. Each of the big teams have cheap options available though be careful to remember that these players are often backups for good reason and a Chelsea side without Cech is not the same Chelsea side that racked up 11 clean sheets last season.

Despite the massive losses of Fabregas (I need a separate post of him later, one of my top five players in Premier League history) and probably Nasri, the Arsenal side is still talented and I'd advise against challenging them with your Liverpool defenders (ie Jose Enrique) this week. A clean sheet wouldn't blow me away but I agree with these stats that better options exist this week.

Attacking Rankings

Rank Team Opp GPG
1 CH WB 2.11
2 MU TH 2.00
3 AV BR 1.82
4 SC @NC 1.76
5 AR LI 1.66
6 NC SC 1.62
7 SU NU 1.45
7 BW MC 1.45
9 SW WA 1.43
10 EV QP 1.42
11 WW FU 1.32
11 MC @BW 1.32
13 FU @WW 1.29
14 BR @AV 1.13
15 NU @SU 1.11
16 QP @EV 1.09
17 WB @CH 1.03
18 WA @SW 0.97
19 LI @AR 0.97
19 TH @MU 0.97

Those top two matchups look fairly tasty and it's going to be difficult to select a captain not from Chelsea or Man Utd this week. Unfortunately that somewhat limits your options as I'm not sure I'm ready to put Nani and Young back in that class until a couple of questions are answered, while Chelsea's lineup is a minefield of potential and substitute tracksuits.

The masses have opted for Torres in their droves this week and based on a sample size of one gameweek, that looks like a shrewd move. However, after last week's goalless effort and Villas-Boas' praise, you have to think Drogba will be back in the team soon, which is an issue in that in brings the value of Malouda and Torres into question, who may both be pushed into unfavourable positions to accommodate the Ivorian legend. Those dead set on bringing in Aguero, could bring in Torres for one or two weeks to take advantage of the favorable fixtures, then make the switch in GW3 or GW4, thus limiting your exposure if Torres reverts to his awful 2010/11 form.

City's trip to Bolton is a tough one to call as the team seems to change so rapidly that any backwards looking data can look out of date in a flash. That said, Bolton are at least a good home side (5 clean sheets last year) while they have the potential to be top-10 unit with a bit more consistency. I would certainly not consider benching the likes of Silva, Yaya and (if you think he will start) Aguero, but the captaincy may be stretch this week.

Interesting (at least to me) side note: Not many will remember, but Aguero has actually played Bolton before (while at Atletico) and I went to the game with a big fan of South American football who told me Aguero would be the best player in the world within a couple of years. Aguero didn't even start and was sent off about 10 minutes after coming on (I think for spitting but apologies if my memory fails me) with Diouf (remember him?) netting the unlikely winner. Good times.

I am still giving Arsenal their own paragraph out of respect for Wenger (who reads this routinely I'm sure) but in reality the only captain option in this side is now Van Persie. If you own him you need to consider giving him the armband, though until we see how they fair without Fabregas I would understand if you brought in Rooney instead. Arsenal (and Van Persie fans alike) may take some solace knowing that the Dutchman averaged 5.6 PPG in games without Fabregas last season, 6.7 PPG without Nasri and 6.5 PPG with neither.

The mid-level teams throw up a couple of good options this week with Bent and Cahill both catching the eye. Bent has a tremendous fixture and has tended to score goals at home in his career but I didn't like the way the side lined up last week. He's proven he can score goals feeding on scraps in the past though and Blackburn's shaky (and injured) defense can cure most offensive ailments. Cahill is a big risk as Everton haven't played yet and have started slow for the past few years. Not to overreact to one game but QPR were simply awful at last week (Danny Gabbidon, really?) and Everton could certainly get a couple of goals here, which probably means a good outing for Cahill. I think I'd stick to ownership rather than captaincy for now though.

I am staying away from too many captain stats for now as while last year might be useful for high level team decisions, I am reluctant to cite Nani's good home form or that someone kills 'weaker' sides as so much changes on an individual level. These will begin to be integrated as trends emerge for this season.

Reader questions
I am pleased to say my the number of readers of this humble blog appear to be increasing and so I'm having a tough time keep tracking back to each post to answer questions. I will therefore endeavour to group common questions together and answer them within each weekly preview. [Apologies for any typos. It's 1:30am and I don't have the energy to proof read]

amtosh - What do you think about sacrificing quality in positions to ensure you have the potential big point scoring players in others? 
Picking your spots on where to spend your cash is obviously a massive part of the fantasy game and every year the market makers do a good job at setting their prices so things are kept fresh. A few seasons ago, anyone with a sniff of competing in their league had Ronaldo and Lampard while Fabregas and Gerrard often hit equal heights when healthy. We' ve had streaks where premium defenders were captain worthy (Man Utd's 8 straight clean sheets spring to mind) while last year it was almost impossible to find a consistent forward with a decent price tag.

At this early junction it looks like forwards are going to be pretty thin and you might need to spend big to stay competitive. Right now, the only player I'm thrilled about owning who costs between 7.0-10.0m is Suarez. This is made easier by the defensive options, which at least for now appear to be manageable. All the premium teams have potentially cheap links and elite prospects like Kompany, Richards and Luiz can be had for a reasonable outlay. I have no idea what's going on in midfield. I can come up with at least some question marks for everyone valued at 10.0m or above whether fitness (Van der Vaart, Gerrard), role (Lampard, Kuyt) or rotation threat (Nani, Young, Malouda).

In short, feel free to spend big in one area but I would not advise aiming for a team rather than a squad. Too many players have already opted for the likes of Newcastle's Ferguson on the basis that he costs 4.0m but injuries and bans can creep up on you and fielding teams of 9 players will quickly put you in a whole. If you stay alert and agile with your selections you should be able to find cheap options who still provide value.

neil hannam  - would you also not class goal keepers as part of the defensive unit?
This one is in reference to the recent cheapest link post. I don't consider 'keepers as part of defensive unit as saves tend to equalise points so that a Al-Habsi playing a team with 7 clean sheets can match the points total of Schwarzer playing on a team with 14. If we had a long term situation where a Lindegaard or Stuart Taylor were playing every week then of course I'm interested but unless than scenario unfolds I am very much happy with my preferred rotation strategy (incidentally the rotation pairs will be updated somewhere around GW8-10 for the new season stats).

Simon G - Can you, or anyone else, explain the popularity of Moses around here?
Moses is getting a lot of attention over at Fantasy Football Scout and while I agree that some may be overreacting a touch, the logic for liking him is solid. The departure of N'Zogbia gives him not only guaranteed minutes but also frees him up to player with the freedom a player of his talents needs. Though he didn't generate any fantasy points this week, he gave the Norwich defense a torrid time and was Wigan's best player by some distance. It's also important to consider his price tag. The budget midfield options have really been slashed this year and 5.0m players with upside are hard to find. In GW1, 23 5.0m or less midfielders played 70 minutes or more. 7 of these play for newly promoted sides. A further 12 of them lined up in defense or in a clearly defined holding role. Only Moses, Diame, Whelan and, to an extent, Delap played a more advanced role for their respective teams. It isn't unbelievable that one of these other options emerges as a reliable starter (I am very interested in corner taking Fox for example) but until we see more Moses looks like a good starter with substantial upside for a great price.

kurbalija - any thoughts on fellaini and barton?
Fellaini is a decent real life player but he shouldn't be a fantasy option for me. One goal and three assist last year in just under half a season suggest a player whose ceiling isn't too far north of 100 points and has  struggled to stay on the field. Barton is very intriguing given his set piece responsibilities and he generally seems to pull the strings of a decent Newcastle side. He still seems to be available though, and as transfer rumours hot up, the chance of him being benched rises too. There's a nice group of options in the 6.0m bracket, most of whom bring more certainty that Barton, though few can match his ceiling (xP 156). I like him as a one week option but I currently have a GW6 alert on him, after which the fixtures improve and we'll know one way or another where he'll be playing until at least the new year.

riiseriiseriise - Who do u guys figure will fill in for cech? 
Hilario was the backup on the weekend and you have to figure he gets the call here. At 4.5m he has a lot of potential but, as I tweeted earlier Cech has conceded just 0.72 G90 over the past 5 seasons compared to Hilario's 1.20. There's a reason Hilario isn't a starter and while the unit is still very good, it is important to remember that with each injury, the value of the teams decreases. The same can be said for Man Utd. It wouldn't be unreasonable for a team like Villa to lineup a back line with the quality of Fabio, Evans, Jones and Smalling but we wouldn't all be going gaga at the prospect of bringing them in for 5.5m. Just because the crest says Chelsea/United doesn't mean your getting their production from past seasons.

RossWB - So who do you think is worth targeting there [Chelsea]? Torres was (a bit surprisingly) favored against Stoke and it certainly seems like he's the primary option there. Drogba? Malouda? Or perhaps a cheaper option like Kalou or Ramires?
Chelsea are a real enigma right now with so much talent but so few fantasy options. Considering how poor he was last year and given the fact Chelsea played out a dull 0-0 with Stoke, it's amazing to think Torres heads this list, but I think he does. I mentioned above that I think Drogba will slide into this side sooner rather than later but he's too expensive to risk having him sat on your bench, or worse, coming on for 10 minutes of garbage time. Lampard and Malouda's price tags look steep when compared with Silva, Nani and Young while the cheaper options don't look safe enough to play to justify their own tags. The defense remains a good option (unless Hilario messes everything up) but until we see Torres and Drogba playing well together, I don't know how to get fantasy coverage over one of the best sides in the league

Baz0819 - Quick question - deciding between Andy Johnson or Holt as 5.5 strikers.
Both deserve monitoring and I'd throw Danny Graham and DJ Campbell into the mix too. Norwich were a very good goalscoring side last year and their fixtures give them the best GPG predictions over the short term. However, I wasn't blown away by any of the new teams last week and for that reason I might lean towards Johnson whose Fulham side don't have great fixtures but they also don't face anyone who terrifies you for the next few weeks.

@erg79 - Who do you think is a safe pick for a 6.0m or less midfielder? Brunt? I know you think that people are overreaching w/ Muamba...
@NikosEllinas - 6mil question (2nd guessing my team): Sinclair or Jarvis? Or someone else (maybe even cheaper)?
I like Brunt, the Wolves trio of O'Hara, Jarvis and Hunt, Rosicky and Danny Murphy as the pick of the 6.0m guys. Brunt and Murphy look like the safest picks and of that pair I still like Brunt given his upside. In the short term, Rosicky should log plenty of minutes for a depleted Arsenal side and while he sometimes played a bit deep at Newcastle, he should be free to roam a bit more when the Gunners play at home. I am definitely not in on Muamba, though I do like him a lot of a player. He has the upside of a homeless man's Vieira. I haven't seen enough of Sinclair to comment fully but against City he was stuck out on the left wing and didn't really live up to the 'midfielder playing up front' tag. He deserves to be on this list but I need to see more before I elevate him ahead of the likes of Brunt.

@Jadles - I'm set in keeping Formica and Nani but should I go with Adams, Yaya, Larsson, Bale, Silva, Dempsey?
I'd go Silva, Yaya, Adams, Bale, Dempsey, Larsson with Silva the clear favorite for me. He's a touch pricey but Bale is a player I plan to target after Spurs get the tricky games out of the way.

@vinnycast27 - Ash Cole/Yaya/T Cahil or Luiz/Silva/Downing? Also is Kranjcar worth bringing in?
I like both groups but I'd go with the second one based on Silva. Depending on your defender situation and your feelings on Aguero, I might be tempted to go Luiz/Silva/Yaya but that might be considered to be too much City for some. At 6.0m Kranjcar's upside value is off the charts and he would probably vault to the top of the aforementioned 6.0m list if we knew he would play. However, I just don't see him getting significant minutes as long as Modric is around, so until he moves, he would be an expensive bench player.

I look forward to seeing your comments/questions below, or over at @plfantasy. Don't get too cute with the free transfers this week and stick to the principles which have served you well in the past (like I'm not on my 3,298th version of team already).