Wednesday, October 31, 2012

Dousing the fire, fanning the flames: Gameweek 9

You know the story by now; each player has been assigned a 'Buy', "Hold' or 'Sell' status. These don't necessarily reflect what I would do, more what I believe to be a reasonable action at this time:

  • Buy - the player can be purchased immediately taking consideration of his playing time, fixtures, injury status etc. This status does not necessarily mean you should buy him now as it depends who else you own, only that it is reasonable to do so.
  • Hold, and/or monitor - if you own the player you're not in a rush to ditch him but there's sufficient risk here to justify potential buyers holding off.
  • Sell - the player has too many risks and should either be sold or not purchased. This does not mean he might not have value in the future, just not right now.

Robin Van Persie 12 points
I was initially reluctant to throw a 'buy' tag on van Persie because (a) the formation issues his arrival caused at Old Trafford and (b) his massive price tag. Issue (a) has been swept aside with Rooney dropping much deeper and van Persie racking up plenty of chances at the apex of a talented United side. Issue (b) still looms large though as while van Persie's start for United has been good, he's still only level with Tevez and a hair above Ba, both of whom can be had for substantially less money.

How we deal with van Persie, his United teammates and indeed their cross town rivals in blue is becoming a key issue this year, so I'll pen a separate post on that in the next couple of days.
Status: Cautious buy depending on your budget

Mikel Arteta 11
Arteta did play as far forward as he has all year this week, possibly thanks to Wilshere coming into the team as that pair can genuinely play as a double pivot, with both capable of pushing forward and hanging back with equal strength. That said, Arteta commands a price tag based on his teammates' ability and his former Merseyside glories and his 8 shots for the year do little to suggest any kind of sustainable success.
Status: Sell

Bryan Ruiz 11
It was a nice goal for Ruiz this week but they don't count for any more points and with pretty decent competition up front for Fulham Ruiz has an up hill battle to even play.
Status: Sell

Gareth Bale 10
The main thing with elite players is Bale is not getting carried away and assuming they should be held forever regardless. It's a select few players in fantasy history who can boast that title and Bale probably isn't in that category just yet. That said, he is one of, if not the best fantasy midfielder in the league right now, leading his peers in total shots by a comfortable distance while also creating plenty of chances. The immediate upcoming fixtures aren't great as Spurs have to travel to City and Arsenal in back to back weeks so if you have a squad you're generally happy with you could make a switch after this week and buy him back for likely the same price. The forecast model doesn't appear to be too concerned though, still placing Spurs tied for 5th over the next 8 gameweeks and thus I'd be minded to just hold Bale and accept the lower ceiling for a couple of weeks.
Status: Buy. If you don't own him you might want to wait until GW13 though this week's Wigan game and a potential price increase might mean you need to act sooner rather than later.

James McArthur 9
I've spoken at length about Wigan's impressive front three but can we add McArthur onto that last of developing Latics? 4 total shots for the year and 9 chances created would say no and I'd agree.
Status: Sell

Mikele Leigertwood 9
Leigertwood is essentially the same player as McArthur for fantasy purposes, this time with 6 total shots and chances created. The conclusion - like the song - remains the same.
Status: Sell

Juan Mata 9
I still trust the process, but my decision to put off signing Mata for fear of his upcoming fixtures was a costly choice. I say 'choice' rather than 'mistake' because I still maintain that if Chelsea played at Arsenal, at Spurs and Man Utd 100 times, the number of outcomes where Mata racked up 30 points would be few and far between. Nevertheless, Mata - and Chelsea - have been extremely impressive and the Spaniard now leads teammate Hazard is all major categories despite having played 100 less minutes. His conversion rates are likely due for some regression, but I'm sure no one thinks they'll be buying a 10+ points per game player, and would be more than happy with the 5-6 P90 his underlying stats supports (that might sound modest but consider that with those totals he would hit 200 points for the year).
Status: Buy, probably a strong one

Clint Dempsey 9
When I looked up Dempsey's playing time I was surprised to see that he's starting every game since coming on a sub for his Spurs debut. Over his five starts, Dempsey has amassed decent underlying stats (11 shots, 7 chances created) but they're not totals to really inspire one to turn your back on Bale or force you to consider doubling up on Spurs midfielders. His playing time and success in the past make him one to monitor but I haven't seen enough yet to make that 9.2m price tag look too tempting.
Status: Monitor

Romelu Lukaku 9
Lukaku's shot statistics are fairly astonishing. His 16 minutes per chance is impressive enough (second only to Suarez) but that gets even more impressive when you consider that they were all inside the box (compared to a 38-18 split for Suarez). After Suarez (22 mins per shot inside the box), the numbers start getting big quickly with Aguero (24), Ba (28) and Berbatov (28) the only other regulars to even get under 30. There's more to being a good forward than taking shots, but considering Lukaku is hitting the target 50% of the time, there's not much more to being a good fantasy forward. Can we rely on his playing time, probably not yet, and as I've said before I'd prefer to a bit late to the party than get their when they're still setting up. He remains however one of the players to monitor as the season progresses.
Status: Monitor

Steven Naismith 9
Everton have a number of good attacking options and I'm not even sure Naismith is one of them.
Status: Sell

Carlos Tevez 9
As noted with van Persie above, City need their own post which will be up soon. I'm not sure Aguero is particularly guaranteed more minutes than Tevez (Shots on Target makes soon interesting points regarding Aguero here) and thus if you want access to this City side Tevez might be the best answer for now. I'm probably still buying Tevez based on his upside in a very good side, though he's far from the locked in asset he looked like he could be to open the year.
Status: Somewhere between a buy and a hold

Demba Ba 8
Ba picked up a slight knock this week and is a doubt for next week, though if I owned him (and I do) I wouldn't be selling him given the fixtures to come (WHU, SWA, @SOU, @STK, WIG) after this week's trip to Liverpool. His underlying stats are as good as anyone's and while you wouldn't bet on his next 15 shots on target to go for 7 goals, a five goal pace over 9 games would still be exceptional for his price tag.
Status: Buy, though obviously consider delaying a week if he looks like he'll miss the trip to Liverpool

Luis Suarez 8
Suarez literally needs his own forecasting model as his shot data continues to be off the charts yet until recently the returns have been moderate. At 9.7m, the natural comparison is Tevez and while I'd still give the latter the edge, the fact that Suarez is pretty much guaranteed to play every week and Liverpool have shown signs of improvement of late makes this decision much closer. We've long pushed Suarez to one side a but because of his rash shots from long distance which don't really translate to goals, but, this year to date he also leads all forwards in shots inside the box by 10 (38 versus Ba's 28). The fixtures aren't great given the away trips to Chelsea, Swansea and Tottenham but Liverpool have actually performed a touch better on the road versus league average than at home so that might not be a huge concern. Liverpool's conversion of shots inside the box into goals has been astonishingly low so far this year (7% versus league average of 15%), so if you believe that the measure will regress to the man at all, they could be well placed for success in the coming weeks.
Status: Buy. I'm not throwing all my weight behind this pick yet but given the uncertainty around the City pair, Rooney and now Ba's injury, Suarez's certainty is looking increasingly attractive.

Wesley Hoolahan 7
Hoolahan is a good player but the stats aren't good enough to justify the risk of him not getting consistent playing time and thus it's impossible to buy in.
Status: Sell

Leon Osman 7
I've mentioned Osman's underlying stats before but they bear mentioning again:

  • Osman: 837 mins, 20 shots, 13 inside the box, 5 on target, 12 chances created
  • Ben Arfa: 835 mins, 17 shots, 4 inside the box, 4 on target, 12 chances created
  • Yaya Toure: 859 mins, 19 shots, 8 inside the box, 7 on target, 17 chances created

It's tempting to cherry pick stats sometimes to tell the story you want but those numbers are legitimately impressive. The fact that he's only 0.6m cheaper than the equally (or more) impressive Pienaar and Fellaini is a blow as they come with more certainty but you also need to balance that with their respective ownership numbers. If take a quick look at Osman's history, we see some very interesting goals scoring trends. If we extrapolate his minutes played to give him 32 starts a season, he would have scored 5 and 6 goals in the past two seasons as well as adding 5 and 10 assists. Those kind of totals suggest a player more than capable of racking up a 120+ point season even if his production didn't improve from last year (and the signs so far are that it will).

Osman is a risk as Pienaar, and particularly Fellaini, have outstanding stats of their own. However, with many managers having already sold their Everton now is a good opportunity to grab Osman for at least the next four games, and roll the dice with a calculated gamble that he can give you some differential returns, all while saving you a buck or two, and possibly even making some.
Status: Buy, as a differentiator

Hal Robson-Kanu 7
Reading's midfield is a mess and Kanu isn't even a regular piece in it. This might be the last time I type his name all year.
Status: Sell

Aaron Ramsey 7
Good team, decent player, so-so stats and a great price tag. Ramsey has taken up some advanced positions of late, particularly this week and with Diaby out his playing time looks as secure as ever (though still not locked down). At 5.6m he's a touch pricey for my taste as a 5th midfielder and I still like other options such as Lallana or Osman as my 4th, but he deserves to be in the conversation, especially given the fact that there too many other ways to reliably access this Arsenal side.
Status: Monitor

Ramires 7
Ramires has substantially worse stats than almost-namesake Ramirez (Gaston) despite the latter having played just 238 minutes this season. When fit, we can consider the Southampton man, the Chelsea version will likely remain irrelevant all year.
Status: Sell

Christian Benteke 7
One of the few players who can match Lukaku's dominant shot totals is Benteke, though he too isn't guaranteed playing time. 21 shots, 19 inside the box and 6 on target all strongly backup his couple of goals and given his limited opportunities, he could have potentially doubled those totals if started every week. If the other bench options were less talented I'd be happier but with Bent lurking on the bench, Benteke is going to face some risk of rotation virtually every week. If you believe that's hyperbole, consider that Grant Holt was only given 2249 minutes under Lambert last year despite notching 15 goals in a great first campaign in the Premier League. The next four games are a nightmare for Villa, but after that they open up nicely so we've got a good trial period to see if the Belgian youngster can hold off Bent.
Status: Monitor

Garath McCleary 6
See Robson-Kanu, Hal

Beausejour 6
Beausejour is essentially just Emerson Boyce without the chance of clean sheets (and actually with worse stats).
Status: Sell

Tom Huddlestone 6
If he was the focal point of a lesser team I could see Huddlestone being a useful fantasy player as he has an excellent eye for a pass (his reverse cross for his assist this week being a good example) and can score from long range. However, at Spurs he's buried behind Sandro and Dembele and thus is unlikely to get regular minutes anytime soon.
Status: Sell

Brett Holman 6
Holman's stats are useful and his playing time looks as guaranteed as anyone in this Villa side. With Benteke, Bent and Agbonlahor to pick from up front, his created chances should be converted at a reasonable rate and he's managed to pickup his shot totals of late. As with Benteke, the next four weeks looks too tough to consider Holman, but he's worth looking at over that period.
Status: Monitor

Papiss Cisse 6
Cisse is a good player and there's something to be said for zigging while other's zag but he has half as many shots as teammate Ba yet still costs 0.5m more. Unless Newcastle's chance creation starts looking very different, Cisse will remain below Ba in our pecking order.
Status: Sell

Dimitar Berbatov 6
We all know how Berbatov is rated in these electronic pages but some water has to be thrown on the flames given his potential rib injury. I'm struggling to find an update beyond what we heard after the game (that the rib might be broken) so that one's worth keeping an eye on. Given the upcoming games my enthusiasm for Fulham is tempered somewhat and so this injury probably pushes Berbatov into the hold rather than 'buy' category. That could change with a green light on the health front, though I'd still have some concern over those fixtures.
Status: Hold and monitor

Jay Rodriguez 6
Rodriguez has some upside and his stats have been useful when he's played. With Ramirez out for another month, Ramirez may well offer some short term value for those really struggling for cash up front, especially given how much the forecast model likes Southampton. I'd prefer to spend the extra 0.8m-1.0m it takes to snag Lallana or Lambert but if you can't afford to take that jump, Rodriguez is a sneaky way to access this team.
Status: Buy if you're desperate for cheap options

Monday, October 29, 2012

Individual forecasts: Historic player data

One of the issues of the new forecast model is deciding which kind of shots to use to forecast player success: all of them, only those in the box or only those on target. Those on target have the best correlation to goals scored, however, there is a slight concern with limiting ourselves to that data alone. Consider the below example:

Robin van Persie: Appearances 10, Total Shots 55, 10 Shots On Target, 3 Goals

If we only look at his shots on target we see that he is averaging just one a game, with 30% of them hitting the back of the net. The issue is that he has historically hit the target at a much better rate than 10/55 (18%), having a success rate more in the 44% range. We therefore need to adjust for the fact that we believe 24 of his next 55 shots (44%) will hit the target and thus his expected goals will be higher.

We have some issue about how to generate this historic rate, especially with regards to what data to use, but for now I'm happy to look at such data for players like van Persie who have a proven history in the Premier League, even if in his case, it was with another team (someone like Berbatov, who has moved to a weaker team might be different).

The next issue is whether we should adjust the rate at which shots on target are converted into goals. This one is much less clear. If you look at the below table, it shows all the significant fantasy players and the rate at which their shots on target have become goals. If you look to the bottom of the table you will note an average conversion rate, and the colouring shows whether players are above/below that average. The keen eye will note that only five players (Balotelli, Berbatov, P Cisse, Fletcher and Nolan) have exceeded league average every year while only four (Bale, Fellaini, Santi Cazorla and Silva) have consistently been below average.



My (admittedly early) conclusion is that there is little reason to think that SoT conversion is consistent year on year and, frankly, is probably a combination of luck, strength of opponent 'keeper, and of course some skill. Issues like where shots are taken from, quality of teammate etc can also be factored in, but for now I'm happy to apply a standard league average rate to all players (probably split by midfielders/forwards once I run the actual numbers). So if you hit the target 10 times, we're going to give you credit for ~3 goals.

We wouldn't want to expand that rate to include defenders of course, or possibly even the league's weaker players, but given that we're really only looking at the top 100 or so attacking players, I'm comfortable that such a generalisation will do more good than harm (are you, for instance, ready to give Cisse a goal for 53% of all his SoTs, when so far this year that rate is 20%). As always I welcome any objections/suggestions below.

Lineup Lessons: Gameweek 9

Arsenal
Mannone, Sagna, Mertesacker, Vermaelen, Andre Santos, Arteta, Wilshere, Cazorla, Ramsey, Podolski, Giroud
Subs: Martinez, Koscielny, Walcott, Coquelin, Arshavin, Jenkinson, Gervinho

Poor old Jenkinson made way for Sagna who has finally shaken off his injuries. Jenkinson did a very good job while deputising for the French full back but for the overall value of the team Sagna probably gives them a decent boost at both ends of the pitch. The downside, of course, is that we lose one of the safe and cheap ways to access this unit. If Mannone can continue to hold of Szczesny when he returns to fitness he would be the obvious pick, though a lot of managers are hesitant to use too many transfers tinkering with 'keepers who might not always play. Mertesacker still faces some competition from Koscielny but he appears to be fairly safe for now and would be my pick for Jenkinson owners looking to move on. Gibbs should be back soon so the left back spot is still up in the air.

Wilshere came into the side and settled in very nicely. His presence, as a player who can genuinely play both ways, should allow Arteta to have more flexibility, and while I don't see either as necessarily viable fantasy options, they do make this team better and should help create more chances for the widely held options like Cazorla and Podolski. On that pair, I believe the panic is a touch extreme as Arsenal have generated 36 shots over the past two 'poor' displays including 14 on target. The narrative is that for all their possession they're not creating chances but the stats simply don't backup that assertion. True, they need to be a bit more 'clinical' but unless you have genuine long term concerns of Cazorla, Podolski etc hitting the target consistently or getting a few of their shots past the opponents 'keepers, I don't see much reason to worry as yet.

Aston Villa
Guzan, Herd, Vlaar, Lowton, Bennett, Albrighton, El Ahmadi, Holman, Delph, Agbonlahor, Benteke
Subs: Given, Ireland, Bent, Westwood, Bannan, Weimann, Lichaj

We're seeing some stability in this lineup but at the key positions it's still very hard to like anyone too much. Bent, Ireland, N'Zogbia, Agbonlahor and now Benteke have all shown enough flashes of talent over recent seasons to justify attention but none of them look guaranteed to even play every week, never mind enjoy fantasy success. Benteke is the latest player to rise and his underlying stats (21 shots, 6 on target) are extremely encouraging and more than backup his couple of goals to date. What Lambert will do next remains unclear and thus it's all but impossible to get too high on Benteke but if you believe Bent is truly finished as a starter here, the Belgian front man at least gives us something to consider from this Villa side.

Chelsea
Cech, Ivanovic, Luiz, Cahill, Cole, Mikel, Ramires, Hazard, Oscar, Mata, Torres
Subs: Turnbull, Romeu, Moses, Marin, Sturridge, Azpilicueta, Bertrand

I liked Chelsea's prospects a few weeks back but I was concerned the fixtures (at TOT, MUN) weren't good enough to buy into their star assets and thought caution was the best approach for a couple of weeks. Chelsea simply scored six goals over the two games with Mata in particular, but also Hazard and Oscar having big contributions. The seven games before their off week don't look outstanding but there isn't much to be scared of either and thus this trio should be catapulted close to, if not atop your transfer target list. We'll get into the data more in fanning the flames but I'd suggest Mata remains the pick of the bunch, followed by the cheaper Oscar who has been a touch unlucky with regards to both shot and assist conversion so far this year.

Everton
Howard, Coleman, Jagielka, Distin, Baines, Naismith, Osman, Neville, Mirallas, Fellaini, Jelavic
Subs: Mucha, Hibbert, Heitinga, Oviedo, Hitzlsperger, Gueye, Vellios

Not much to report here in terms of personnel with the bigger stories being the return of Fellaini and the absence through suspension of Pienaar (who should be back next week). The goals came from a couple of less heralded options though Osman is a player I've highlighted before and he still offers potential given his 0.6m discount over his pricier teammates. I'd probably try and find that extra cash personally, but if you're budget is tight and you want some access to this talented Everton team, don't dismiss Osman without checking out his stats.

Fulham
Schwarzer, Riether, Hughes, Hangeland, Riise, Duff, Baird, Diarra, Richardson, Rodallega, Berbatov
Subs: Stockdale, Senderos, Sidwell, Petric, Ruiz, Karagounis, Dejagah

Fulham seemed to play a more conventional 4-4-2 here with Rodallega given the nod up top which meant no place for either Petric or Ruiz. Richardson and Duff took up some very advanced positions on the wings and remain viable options, though neithers' underlying stats have really captured the imagination to date. The biggest story from the game is probably the rib injury to Berbatov which could see him miss a couple of games - despite playing on, and scoring, in this game.

Liverpool
Jones, Wisdom, Skrtel, Agger, Jose Enrique, Fernandez Saez, Sahin, Gerrard, Allen, Sterling, Suarez
Subs: Reina, Assaidi, Henderson, Coates, Downing, Carragher, Shelvey

I joked last week about Jones keeping Reina out of the side after last week's clean sheet and then this week's team sheet came in and I was momentarily stunned. Rodgers confirmed however that Reina hadn't trained much all week and was held out purely as a precaution. Enrique also came in because of an injury to Johnson but again the long term impact of that move is limited. Otherwise it's as you were with Sahin, Saez and Sterling apparently cemented in this side alongside Gerrard and Suarez.

Man City
Hart, Richards, Nastasic, Kompany, Clichy, Nasri, Toure, Barry, Kolarov, Tevez, Aguero
Subs: Pantilimon, Lescott, Toure, Evans, Sinclair, Dzeko, Balotelli

Despite Mancini's suggestion of possibly benching Richards for daring to criticize his very-criticizable tactics, no such change was made but Richards was then stretchered off and looks like he could be out for a while. With Zabaleta and Maicon also out we could possibly see a prolonged move to the 3-5-2, though Mancini could also opt to slide Kolo Toure in at right back. It's tempting to simply walk away from this whole unit but they once again held the opponents to very few shots on target (2) and their potential remains as high as anyone.

Offensively Kolarov got a somewhat surprising start on the left wing, but he was also forced to leave the game early. A long term injury to him would make Clichy a lot more attractive at the back, but his absence does little to solve the issues up front. At least Mancini showed a willingness to play Tevez and Aguero together but so long as Dzeko and Balotelli are lurking on the bench and the boss won't contemplate playing three at once, it's going to be a headache for fantasy managers. Yaya was once again played quite deeply and his leash seems to be much shorter than at times last year. He is probably overpriced under the current situation.

Man Utd
De Gea, Da Silva, Ferdinand, Evans, Evra, Valencia, Carrick, Cleverley, Young, Rooney, van Persie
Subs: Lindegaard, Anderson, Giggs, Hernandez, Nani, Welbeck, Scholes

United came out in what looked like a more classic 4-4-2 with Young and Valencia out wide to support Rooney and van Persie but for large stretches of the game Rooney played very deep to link up the play forcing the formation more into line with the en vogue 4-2-3-1. Though he was unlucky not to get an assist for the Luiz own goal, Rooney didn't put up the kind of numbers that support his price. Now, this might only be the case against the league's better teams but if you're paying ~12m for a player, you need him to at least be a threat every week, and ideally you want him to captainable in every game. I will expand on United's prospects as we focus on the new long term forecast but my snap judgement is that van Persie remains the only real player I'm considering from Old Trafford on the attacking side of the field.

Newcastle
Krul, Simpson, Coloccini, Williamson, Santon, Ben Arfa, Cabaye, Perch, Gutierrez, Ba, Shola Ameobi
Subs: Harper, Anita, Cisse, Obertan, Steven Taylor, Sammy Ameobi, Ferguson

Unfortunately the headline here is once again negative with goalscorer Demba Ba departing through injury. It didn't appear to be too bad so his absence could be short, though with Cisse waiting in the wings, at least a precautionary rotation looks likely. Steven Taylor was on the bench, but was presumably not quite fit enough to start again. I assumed Taylor would immediately get back into this team when healthy so that's a situation to monitor next week if he remains absent from the starting eleven.

Norwich
Ruddy, R Martin, Bassong, Turner, Garrido, E Bennett, Tettey, Johnson, Pilkington, Hoolahan, Holt
Subs: Rudd, Snodgrass, Howson, Jackson, Morison, Barnett, R Bennett

Hold the presses: an unchanged side in Norwich! Hoolahan put up decent numbers again and now he appears locked into the first team he offers some value at 5.5m. Snodgrass led the team in shots despite despite playing just half an hour but Chris Hughton and co still remain unconvinced he can contribute to a side who've already racked up a whopping 7 goals in just 9 games.

QPR
Julio Cesar, Bosingwa, Mbia, Nelsen, Traore, Wright-Phillips, Diakite, Granero, Hoilett, Taarabt, Zamora
Subs: Green, Ferdinand, Cisse, Mackie, Onuoha, Ephraim, Faurlin

Granted, they almost held Arsenal goalless this week in what would have been an excellent draw at the Emirates, but they still surrendered 22 shots (12 inside the box, 9 on target) and managed just 6 (3, 2) of their own. Granero, Hoilett and Taarabt have all earned more consistent playing time and shown decent  underlying stats over the past month or so and there's certainly potential within this group. However, this game should serve as a reminder that despite the promising names on the teamsheet, this is a still a team whose options are likely only ever going to be good rotation options rather than every week starters.

Defensively, they're still giving up over 10 shots inside the box per game, and that rate hasn't improved over the more recent games. Even if the lineup settles down I don't see any value there unless Hughes can improve their play and/or tactics.

Reading
McCarthy, Cummings, Gorkss, Mariappa, Shorey, Tabb, Leigertwood, Kebe, McAnuff, Pogrebnyak, Roberts
Subs: Federici, Gunter, Pearce, Le Fondre, Hunt, McCleary, Robson-Kanu

It was good to see Reading get three goals this week, even if the sources of points were widely spread (no player scored or assisted more than a single goal). The key players in this lineup are fairly secure with some minimal rotation in midfield, yet outside of McAnuff and Pogrebnyak who've been decent this year, no one else has really emerged a viable option.

Southampton
Boruc, Clyne, Yoshida, Fonte, Fox, Puncheon, Schneiderlin, S Davis, Lallana, Lambert, Rodriguez
Subs: K Davis, Hooiveld, Ward-Prowse, Do Prado, Mayuka, Chaplow, Reeves

Lambert returned to the starting eleven here, though last week's rotation still looms large in my mind when considering placing a 'buy' tag on the front man. Southampton forecast very well in the new model and their opponents in the next seven gameweeks wouldn't scare anyone. The front six looks fairly secure but given their respective roles and the uncertainty around Lambert, I'm not sure I can get behind anyone other than Lallana right now.

Stoke
Begovic, Cameron, Shawcross, Huth, Wilson, Kightly, Whitehead, Adam, Nzonzi, Walters, Crouch
Subs: Sorensen, Jones, Owen, Upson, Etherington, Wilkinson, Palacios

I liked Marc Wilson a lot so it was a blow to see him stretchered off, though in Wilkinson we have an equally priced player who should take over that job full time in Wilson's absence, without overly compromising the unit's value. Stoke's defensive fixtures remain excellent so even if you're worried about Wilkinson' security, you should really be looking at someone from this team to boost your budget defensive line.

I like the attacking options from this side on paper, though they haven't created a ton of chances and at a certain point that has to become a concern. The good news though is that those chances are concentrated with just a handful of players so likes of Kightly, Walters and Crouch remain ownable. The upcoming games are solid, especially the next six at home, where Stoke have traditionally fared very well. Those suffering with injuries to Berbatov or Ba (if confirmed to be long term) could do a lot worse than take a look at Crouch for a consistent and affordable option up top.

Sunderland
Mignolet, Gardner, O'Shea, Cuellar, Rose, Larsson, Cattermole, Colback, Johnson, McClean, Fletcher
Subs: Westwood, Bardsley, Campbell, Kilgallon, Vaughan, Saha, Sessegnon

Sessegnon being dropped was a bit surprise to me, as while I understand the need to pull back the attacking options when away from home, I'm surprised it was Sessegnon who suffered (as appose to benching Johnson and shifting Larsson back out wide, for example). That said, as good as he was last year, Sessegnon just doesn't have the stats to backup anything close to his price tag and the risk of rotation should be the final nail in his coffin for anyone who still owns him. The rest of the side looks pretty settled, though the return to fitness of Bardsley (who came on a sub here) could spell the end of Gardner's place in the side. Despite his moderate attacking upside, Cuellar will likely remain the value pick on this back line.

Swansea
Vorm, Rangel, Chico, Williams, Davies, Ki, Routledge, Britton, de Guzman, Hernandez, Michu
Subs: Tremmel, Dyer, Graham, Monk, Shechter, Tiendalli, Agustien

The injury to Vorm is a big blow as he's one of the better keepers in the league and is widely held by some 17% of managers. For me, that number was way too high as Swansea just haven't looked as solid at the back as they were last year, but if you've transfers to burn, Tremmel at 4.0m would instantly be the best minimum-priced 'keeper around. I'm not crazy about the fixtures so I wouldn't be rushing to make such a move but it might make sense if money is tight and you're currently stuck with someone even worse (anyone from Southampton or Reading).

Offensively Dyer and Graham's lack of playing time has reached a tipping point where they cannot be owned until they prove they can reestablish their respective places. That leaves Michu, Ki and Routledge along with the potential of Hernandez as possible options with the unheralded Ki actually enjoying the best stats of late among that group. At 6.0m and with zero non-appearance points to date it's tough to sell Ki to Michu owners but the underlying stats are great and 1.3m saved is worth thinking about.

Tottenham
Friedel, Walker, Caulker, Gallas, Vertonghen, Sandro, Huddlestone, Lennon, Dempsey, Bale, Defoe
Subs: Lloris, Naughton, Dawson, Sigurdsson, Falque, Livermore, Townsend

No real surprises here as it's looking increasingly certain that Dempsey has displaced Sigurdsson as the team's starting AMC now and had a useful game here both in statistical terms and of course with his goal. Despite always being on the lookout for differentiators, I find it impossible to contemplate grabbing Dempsey over Bale given the small difference in price, and if I wanted to do something different I'd look to Lennon, who had a better game based on the eyeball test than the stat sheet would suggest.

West Brom
Foster, Tamas, McAuley, Olsson, Ridgewell, Morrison, Gera, Mulumbu, Dorrans, Odemwingie, Lukaku
Subs: Luke Daniels, Popov, Rosenberg, Long, El Ghanassy, Jara Reyes, Fortune

Three shots, all on target and a goal for Lukaku add a bit more weight to the assertion that he really should be starting over Long if West Brom hope to push on a bit this year. The potential partnership with Odemwingie is very promising and at 6.4m Lukaku could make for an interesting differentiator if he can win Clarke's trust and start a few games in a row. There's more to being a good forward than simply shots and goals of course, but consider that Lukaku now has 20 shots (10 on target) and three goals compared to say Torres (18-7-4) despite playing just over a third as many minutes and his upside becomes evident. It's too early to bring him in without further proof of his starting, but the goal will likely help his cause.

West Ham
Jaaskelainen, Tomkins, Collins, Reid, McCartney, Jarvis, Noble, Nolan, Diame, Benayoun, Carroll
Subs: Spiegel, Cole, Maiga, O'Brien, O'Neil, Chambers, Hall

West Ham have provided some good fantasy value to date and while no new issues have really arisen with their starting eleven, the fixtures now get pretty tough for a while and thus I wouldn't be rushing to bring any of these guys in. Would the fixtures lead me to selling someone like Nolan, maybe not, though I would consider it if not sat on a profit and if I had free transfers to spare.

Wigan
Al Habsi, Ramis, Caldwell, Figueroa, Boyce, McCarthy, McArthur, Beausejour, Kone, Di Santo, Maloney
Subs: Pollitt, Jones, Watson, Gomez, McManaman, Boselli, Stam

Wigan have one of the more consistent teams in the league, with the added bonus of essentially all their fantasy value coming from that front trio. They only contributed a single assist this week but still accounted for 9/16 of Wigan's shots and can generally be counted on to prosper whenever Wigan do as a team. Wigan have quietly become a good source of fantasy options and should be monitored closely given that they have a number of decent games on the horizon. 

Friday, October 26, 2012

Medium Term Goal Forecast

The mechanics of the new model were expanded on here, so check that out specific questions on how the below forecasts differ from the old goals based system (clue: the differences are varied and significant!). While the weekly forecast is more related to line-up decisions, the below chart should hopefully help with transfer decisions and longer term planning.

Attacking Forecast

Defensive Forecast
Weekly Forecast Predicted goals per game scored for the next eight gameweeks, based on shot data and team conversion rates
Avg Average goals per game forecast for the next eight gameweeks
Gr The percentage of 'green light' games over the next eight gameweeks, currently defined as a forecast goal tally equal or greater than two for attacking sides and less than one for defensive teams.
Am The percentage of 'amber' games over the next eight gameweeks, currently defined as a forecast goal tally between 1 and 2.
Rd The percentage of 'red light' games over the next eight gameweeks, currently defined as a forecast goal tally less than or equal to 1 for attacking teams or more than two for defensive teams.

Gameweek 9 Preview

This week's preview piece is going to be in two pieces. The first piece will be the actual forecast for the week and a very high level explanation of what the numbers mean. Lower down you'll find the mechanics behind the forecast along with some identified weaknesses and some proposed questions as to where we go next with the model.

First then, here is this week's forecast:


xG Shots Expected goals based on shot data for both the highlighted team and their opponent to date
xG Shots Regressed Expected goals based on shot data for both teams, this team regressed using league average shot conversion rates.

The Nuts and Bolts
This is the end result of a project I've been working on for a couple of weeks now, and it was discussed at length in an earlier piece. Now the data has actually come together though, I wanted to take the opportunity to run through an example so as to (a) help everyone understand where I'm coming from and (b) I find that explaining something to someone else helps you identify issues or errors in the logic. As I've said before, this is still a work in progress and I'm well aware that there are still issues to resolve, some of which I'll address below. I welcome any suggestions below, on Facebook or on Twitter.

Given my love affair with Arsene Wenger, and the fact that Arsenal generate elite numbers which should illustrate some ideas nicely, let's use the example of Arsenal welcoming QPR to the Emirates as our example:

The first step is to try and forecast the number of shots a given team will have this week, split between those taken inside and outside the box. There are a number ways one could do this but where I've started is to look at how a team has done against opponents to date compared to the league average against that team. An example of this can be found in table 3 here. Though it's tough to analyse this data in a vacuum, I will include the full tables below to see exactly what I'm working with. The first table shows the average rate which each team is exceeding/trailing the league average in terms of generating shots while the second table shows the same trends on the defensive side of the ball (note that in the second table a negative number is a good thing):


As we can see, Arsenal's attacking numbers are a bit odd as they've outperformed the league average in every category apart from shots inside the box at home which they've been slightly below average at. I would suggest this is purely due to small sample size issues rather than an underlying trend but that kind of trend could be problematic were it to continue.

We can see from the second table that QPR give up 12 shots inside the box (SiB) away from home and thus if we apply Arsenal's premium/discount of -4% we get a forecast of 11. We then look at how many SiB Arsenal are averaging at home (12) and apply their opponents premium/discount (16%) for how many SiB they are giving up away from home, arriving at a forecast of 14. For now, I then take a simple average of these two amounts to get the expected SiB for Arsenal, though this might need tweaking if we can establish whether a team's attack or their opponent's defense has more impact on this number.

The same logic is then followed for shots outside the box (SoB) until we have an expected number of total shots for Arsenal (19) split between those in (13) and out (6) of the box.

The next part of the calculation is where the divergence in the first two graphs stems from. The first graph is derived using Arsenal's shot conversion rates i.e. the rate at which they convert shots into goals, again split between those inside and outside the box. Once again, these numbers are a bit tough to read out of context, but I'll include them below again for reference:


So now we simply take Arsenal's expected SiB (13) and apply the above conversion rate (12%) to get 1.5 expected goals. We then perform the same calculation for SoB (6 x 6%) to get 0.4 goals giving us a total expectation of 1.9 goals for the week.

The point of the second graph is to eliminate the potential volatility of these conversion rates by using league averages. You can make the argument that some, or indeed most, of these rates make sense with top teams like Man Utd and Chelsea ranking well while teams with poor or inconsistent strikers coming in towards the low end of the range. However, after just eight weeks there's clearly going to be some significant movement to come, and ignoring it completely leads to the somewhat perverse ranking this week which sees Chelsea come out as the best play of the week despite facing a good opponent, largely because of their incredible 24% conversion rate of SiB.

The second graph therefore uses the league average rate of 15% for SiB and 4% for SoB which in Arsenal's case will actually improve their forecast (remember, despite the terminology being a bit confusing compared to how we usually use the word, regression to the mean can be better thought of like conversion rather than regression in a negative sense). In truth I think the best answer probably lies somewhere in the middle of the two graphs so I'd be minded to check both before making any key decisions. The extent to which teams' conversion rates need to regressed and when they stabilise is a project I hope to undertake soon.

So there it is. As I say, I welcome any suggestions and the next step is to get involved in Shots On Targets' analytics project to help take these models to the next level. I am almost finished with the individual player model too so hopefully we'll have some captain data by this time next week too. I'm currently in the final skirmishes of that battle, deciding how to calculate an individual player's historic conversion rate (for example, how do we treat Demba Ba's data from his time at West Ham? How about his seasons at Hoffenheim?). Input and suggestions on that model are also welcome, though they may be best saved for next week.

As always, thanks for reading and for your patience in helping me put together these next generation models. I've had a busy few weeks so the standard fantasy pieces like reader questions have fallen by the way side a bit, but I hope to resume that service very soon.

Thursday, October 25, 2012

Dousing the fire, fanning the flames: Gameweek 8

You know the story by now; each player has been assigned a 'Buy', "Hold' or 'Sell' status. These don't necessarily reflect what I would do, more what I believe to be a reasonable action at this time:
  • Buy - the player can be purchased immediately taking consideration of his playing time, fixtures, injury status etc. This status does not necessarily mean you should buy him now as it depends who else you own, only that it is reasonable to do so.
  • Hold, and/or monitor - if you own the player you're not in a rush to ditch him but there's sufficient risk here to justify potential buyers holding off.
  • Sell - the player has too many risks and should either be sold or not purchased. This does not mean he might not have value in the future, just not right now.
Juan Mata 18 points
A couple of us on Twitter have been discussing Mata a lot these past couple of weeks (h/t to TalkOfTheCrowd for the original tip) and it seems I might have missed the boat both in terms of signing him for my own team and to talking him up more in these electronic pages. His underlying stats have been tremendous since those first few games where he played in a much deeper role, indeed over the last four gameweeks:

Mata: 361 minutes, 13 total shots, 6 on target, 3 goals, 13 chances created, 5 assists
Oscar: 344 minutes, 12 total shots, 5 on target, 0 goals, 8 chances created, 1 assist
Hazard: 339 minutes, 5 total shots, 3 on target, 1 goal, 6 chances created, 0 assists

Mata has obviously benefited from a touch of luck over these games with his shots and chances created producing points at an unsustainable rate, but however you slice it, he's clearly the better Chelsea player at the moment. If they were the same price I'm not sure who I'd pick but given the fact that Mata is (a) cheaper and (b) is somewhat of an elite differentiator (well 10% vs 42%) he is the clear pick for me and would have been the name to replace Silva if it wasn't for his fixtures. He's still very ownable but MUN, @SWA, LIV, @WBA, MCI doesn't promise huge returns so his value takes a knock. That said, he is at least all but guaranteed to play every week so he still might be one of the top three elite midfield elite options.
Status: Buy, though do check those fixtures and the upcoming long term model forecast before going overboard.

Chris Baird 15
This was Baird's first goal since about 1923 and came on his first shot on target for the season. The fact that he is classified as a defender but is playing further up field is intriguing but when everyone is fit I'm not sure he's a first team pick and so I'd prefer to pay the extra 0.1m and get the ever-reliable Hughes.
Status: Sell

Mark Noble 15
I wasn't overly sold on Noble coming into the year but his penalty duties do make him somewhat interesting given his price tag. Other than those chances from the spot, Noble offers little by way of goal potential, though his 17 chances created suggest he should be able to rack up a couple of assists for the year. West Ham have a good attacking unit so future penalty chances should come Noble's way and at 5.0m that shouldn't be ignored. Personally I still like the upside of Maloney and Sterling but Noble is very much ownable for the time being.
Status: Buy, though for me, better options exist

Wayne Rooney 14
I'm actually going to defer here as I have a separate piece due tomorrow based on the new model forecast, in which United look simply irresistible. I've struggled with what to do with van Persie and Rooney to date but their upcoming games force the issue somewhat so I'll dedicate a bit more time to this important issue.
Status: Procrastinate

Kevin Nolan 12
Yup, Nolan is legit. Not much more to add that I haven't waxed lyrical about before. He's getting plenty of chances and is one of the best finishers in the league at his position. The arrival of Carroll only strengthens his upside and so Nolan remains one of the very best mid range options around.
Status: Buy

Raheem Sterling 11
Sterling was highlighted a couple of weeks ago in the FFS moneyball series and nothing has changed to suggest his value is diminished. Given his price tag and apparent fixture in a developing side, he's about as good as it gets in the ~5.0m range.
Status: Buy

Robin van Persie 11
See Rooney, Wayne

Jonathan De Guzman 11
De Guzman has the dubious combination of a lack of guaranteed playing time and a deeper role when he does get the nod.
Status: Pass

Edin Dzeko 11
Among players with more than a handful of minutes, van Persie has the second best minutes per goal rate at 103; Dzeko's is 61. Despite his double here though, I still don't see him playing often enough to rack up enough total points to help you, even if his P90 is off the charts.
Status: Sell, until someone gets injured or retires to Argentina to play golf

Jon Arne Riise 10
See Hughes, Aaron below
Status: Sell

Danny Welbeck 10
Welbeck's in a similar position to Dzeko in that his underlying stats are useful but playing time is just far too sparse to consider him. Great for United fans and Ferguson, not so much for fantasy managers.
Status: Sell

David Hoilett 10
We've been waiting for someone from this QPR side to emerge as a legitimately ownable option, and while I've flirted with Granero and even Taarabt, Hoilett might finally be the answer. Among midfielders with a minimum 400 minutes, Hoilett trails just Fellaini, Cazorla and Bale with an attempt on goal every 28 minutes and his 5 shots on target rank alongside or ahead of players like Gerrard, Silva and Ben Arfa. The knock on Hoilett is that with three chances created he's essentially a one category player and he's therefore immediately going to be compared to Nolan, who as of now is clearly a better option. Of course, Nolan is 0.7m more expensive and is owned by 16% more managers, but once you make the comparison you see the uphill battle facing Hoilett. If he can lock a first team place down to the point where he's as likely to play as someone like Nolan I'll be very interested in his services. Until then, I feel his status is stuck on monitor.
Status: Monitor

Gary Cahill 9
In theory Cahill could be a nice short term fix in Terry's absence given his price (6.3m) and goal threat but with the next four games offering limited chances of clean sheets, I don't believe Cahill is worth looking at, at least until Terry finally gets incarcerated for good.
Status: Sell

Eden Hazard 9
I touched on Hazard above with regards to Mata, so there's not much to add here. Hazard is certainly good enough to own in a vacuum, but given his inflated price and ownership numbers I'd be looking to sell sooner rather than later if I owned him. He's still a hold though.
Status: Hold

Javier Hernandez 9
Let me know when he signs for a team where's he's higher than fourth on the depth chart.
Status: Sell

Yohan Cabaye 9
I really like Cabaye as a player and his underlying stats suggest a player who still deserves some attention. He doesn't stand out in any one major indicator of future success but he has enough shots and chances created to suggest he can contribute in all categories this year and given that his numbers are essentially the same as Ben Arfa so far, he might offer a nice way for Ben Arfa's owners to free up ~1.0m in cash. The next seven games look pretty outstanding for Newcastle and while I don't see Cabaye as the kind of player to rack up huge points, I like him over the next 6 weeks or so and one can at least make an argument that he represents a good alternative to owners looking to cut loose with Michu before the coming regression.
Status: Buy

Brad Jones 8
*Shock jock radio voice* Here's what no one's talking about: Brad Jones has as many clean sheets as Pepe Reina this year. Look, this is why you need an English 'keeper in the Barclays Premier League. Jones' mental toughness is clearly the difference and Reina should be sold. Can you even name one successful foreign keeper in Barclays Premier League history? *End radio voice* So, yeh, I'm aware that Jones is Australian and that he's very much not Reina's replacement. Pass.
Status: Sell

Michu 8
Oh Michu, how I want to sing your praises but you keep succeeding at such a ridiculous rate that there's no way I forecast any kind of improvement, nor even sustained production at this pace. The good news is that his 20 shots are elite and are indicative of future success for sure. It's just that I fear we're now paying a premium for a player who has scored 6 times from just 9 shots on goal. At 7.3m he's still affordable but it's impossible for me to take him over Nolan at 0.8m less, and I'd have to think long and hard about players like Morrison (1.2m less) if there's anywhere else in my team to improve.
Status: Hold

Emerson Boyce 8
I touched on Boyce in last week's moneyball piece on Marc Wilson and the goal here does nothing to change my view. His 9 shots and 8 chances created are a great bonus for a defender but when you have very little chance of a clean sheet every week, you need to be comparing those totals to midfielders as offensive points are really your only source of production. I believe you can find value for 4.5m in defense much easier than in midfield so you're better served going with a combination of, say, Aaron Hughes and Shaun Maloney than Emerson Boyce and any of the dubious options in midfield for 4.5m.
Status: Sell

Adam Lallana 8
Despite my private joy of Lallana bailing out my own team this week, I felt a twinge of guilt for not talking Lallana up more these past few weeks as he really deserves more attention. 14 shots along with 16 chances created put him in a fairly elite group of players who've racked up 14 in each category, alongside Cazorla, Gerrard, Hazard, Bale, Mata, Yaya, Pienaar and Morrison. Along with Morrison (a player I have hyped a lot on here), he obviously represents a huge discount to his peers and thus deserves a strong look for anyone needing a excellent fourth midfielder or even a good number three. I won't get into the detail now, but Southampton rank very well in the upcoming long term forecast based on the new model and now is as good a time as any to grab Lallana. With Lambert suffering from rotation again this week, Lallana looks like the clear favourite to capitalise on Southampton's favourable fixtures which last right through until their off week in GW17.
Status: Buy

Alexander Tettey 8
Perhaps Tettey, still a youngster, might develop into a player to monitor, or perhaps he'll fade away into obscurity never to be heard from again. Whatever his path, he'll tread it far from our fantasy teams . . . at least for now.
Status: Sell

Grant Holt 8
Holt's a useful player and is probably playing right around his potential at the moment. I'd like to see him get a few more chances to really excite me but he is a good finisher who has a proven ability to convert shots at an above-league average rate. His 11 chances created are a bonus for a player sometimes categorised as an 'old fashioned' striker and thus he has the potential to contribute in every category, which is more than can be said for some of his similarly priced peers. I'm still nervous about the way he's been rotated this year and thus it's tough to put him alongside the likes of Berbatov, but he certainly deserves watching over the coming weeks.
Status: Monitor

Aaron Hughes 7
What does Aaron Hughes need to do to get some love around here? Consider:


Minutes
Total Shots
Chances Created
Hughes
757
4
4
Hangeland
757
3
3
Riise
659
4
7
Riether
757
3
6

Riise is obviously a better attacking threat on paper and Hangeland has shown some ability to get on the end of set pieces in the past, but you're paying between 0.5m and 1.1m more for the chance at those points and I haven't seen anywhere near enough from Hughes' teammates to consider the upgrade. I'm not crazy about the upcoming games in the short term but between GW12 and GW22 Fulham only have a couple of games where you'd be scared to start their defenders, making Hughes a useful 4th or 5th defender as a long term investment.
Status: Monitor, with a view to buying in a few weeks. If you already own him I'd be okay holding on through these two tricky games

Carlos Tevez 7
I'm not sure what to say about this City team. Aguero and Tevez have now only started together once and while they've sometimes had success off the bench, spending 9-11m on non starters is simply not a situation you want to be in. Tevez was on the bench again against Ajax today which probably bodes well for the weekend and he's still just about playing enough to justify his price tag. However, there's no way Aguero is rested so much all year and with Dzeko actually leading the team in goals, it's tough to see Tevez getting guaranteed minutes for the long term.
Status: Hold, just about, but start making contingency plans for when the wind changes and someone else is Mancini's flavour of the month.

John Ruddy 7
I was on the fence about Norwich's defensive prospects coming into the year and so I was willing to give them time for all the new faces to settle. Well we're 8 weeks in now and while the results haven't been terrible (two clean sheets to date) they rank in the bottom half in shots conceded in the box and are dead last in the metric over the last four gameweeks. With a team like Norwich you're really buying them for their home games, in the hope they can be rotated with another cheap option and so the fact that they get United, City and Chelsea in the next six games at Carrow Road effectively ends my interest here.
Status: Sell

Sebastien Bassong 7
See Ruddy, John
Status: Sell

John O'Shea 7
I was fairly high on this defense coming into the season and while they've been okay, the number of shots they're surrendering is alarming (second worst in shots allowed inside the box). Looking closer though, we see home hope for O'Neil's side. They've faced a brutal away schedule to date (Arsenal, Swansea, West Ham, Man City) and that's where the majority of their shots have been conceded. Indeed, at home they actually rank in the middle of the pack, and have conceded just twice in three games. Given that they're probably therefore only really useful in a rotation strategy O'Shea is a bit more expensive than you'd like, thus I'd prefer Carlos Cuellar here.
Status: Sell, and grab Cuellar instead.

Jermaine Defoe 6
Defoe continues to surprise (me at least) not so much in his success on the field (he's always been a good finished) but that he's kept his place over the more heralded Adebayor. I think it was Einstein who said that the definition of insanity was repeating the same mistake over and over, so I'm not going to say that I'm down on Defoe again, though I'm still struggling to get excited about buying him given the rotation threat he faces. That said, it's probably no more than many other elite players in the league and Defoe's production has very much been in their class. He's second in total shots and first in shots on target and while Spurs' fixture list is up and down, there's enough value there to make him viable.
Status: Buy, though I still have some reservations

Mark Schwarzer 6
As we noted for Hughes, I'm not crazy about Fulham's short term prospects given the fixture list but after GW12 they look like a good buy prospect and Schwarzer can form a very useful rotation strategy with another budget 'keeper (I will do an updated rotation piece based on the new model soon).
Status: Hold for now, becoming a buy from GW12 onwards

Brede Hangeland 6
I like Hangeland quite a lot but as noted for Hughes, I don't see enough to justify the extra cost in bringing him in unless he starts racking up significantly higher shot totals.
Status: Sell, and stick with Hughes

Sascha Riether 6
See Hangeland, Brede
Status: Sell

Nathan Baker 6
Villa's defense is actually a lot better than I expected if you look at the stats but this isn't really a unit to dream on right now and Baker probably wouldn't be the right man even if it was. They get City, United and Arsenal in the next five so this certainly isn't the time to be flirting with any Villa defenders.
Status: Sell

Glen Johnson 6
It's not a particularly telling metric but it says something nonetheless that Liverpool have conceded 12 goals despite holding opponents to a league best 38 shots inside the box, while on league average rates they would have conceded just 6.6 goals. I think we can still conclude that Reina is an above-average 'keeper so I'm not sure what to make of the stats. Perhaps the defense is giving up better shots that other teams? Perhaps it's just bad luck. We need to see more detail before predicting strong regression but given their consistent ability to limit opponent chances (possibly due to their good possession totals) I'd be happy to suggest Liverpool will at least get a bit better at the back and perhaps we've started to see this based on two straight clean sheets. Unfortunately the upcoming games don't look great though on the bright side we have another 5 or 6 games to assess what this defense will offer this year, before the fixtures open up nicely from GW15 on.
Status: Monitor Liverpool, and particularly Johnson's attacking potential to see if he could justify the premium if/when Liverpool's defense starts to improve.

Martin Skrtel 6
See Johnson, Glen
Status: Monitor

Andre Wisdom 6
See Johnson, Glen
Status: Monitor

Michael Kightly 6
13 shots looks promising enough but 5 chances created isn't going to get it done, even for a player with Kightly's low price tag. He's worth monitoring but I don't see enough here to warrant selecting him over the ~6.0m group like Lallana or Nolan, or the cheap group in Maloney and Sterling.
Status: Monitor

Russell Martin 6
See Ruddy, John
Status: Sell


Javier Garrido 6
See Ruddy, John
Status: Sell

Wednesday, October 24, 2012

Lineup Lessons: Gameweek 8

I didn't see a ton of football this week as I was away for the weekend so this week's lineup lessons will be somewhat condensed, focusing purely on surprise omissions and potential issues, and avoiding the often rambling detours these electronic pages have become synonymous with. Indeed, the reader may well find themselves wishing your humble blogger watched less weekly action to keep future posts this short.

Arsenal
Mannone, Jenkinson, Mertesacker, Vermaelen, Andre Santos, Ramsey, Arteta, Cazorla, Gervinho, Giroud, Podolski
Subs: Martinez, Wilshere, Oxlade-Chamberlain, Djourou, Coquelin, Arshavin, Gnabry

Nothing to report here at all. Wilshere made a return to the bench from injury (delighted for him: one of the English players with truly elite upside in my humble opinion), but from a fantasy perspective we're along from that mattering. This was an odd result for Arsenal, though not totally out of character with their recent struggles, and Cazorla and Podolski owners shouldn't be overly concerned.

Aston Villa
Guzan, Lowton, Vlaar, Baker, Bennett, Ireland, El Ahmadi, Delph, Holman, Bent, Agbonlahor
Subs: Given, N'Zogbia, Albrighton, Benteke, Bannan, Weimann, Lichaj

Playing Ireland and Bent seems like a wonderfully novel idea but despite their past successes neither are ownable for now. Indeed, there's very little to like in this whole team, especially with all the players with a modicum of potential being rotated weekly.

Chelsea
Cech, Cole, Cahill, Luiz, Ivanovic, Mikel, Ramires, Hazard, Mata, Oscar, Torres
Subs: Turnbull, Azipilicueta, Bertrand, Romeu, Lampard, Moses, Sturridge

Lampard’s move to a more defensive role has been a godsend for fantasy owners with Mata, Hazard and even Oscar presenting themselves as legitimate every week starters. All three contributed this week, with their respective scores falling in line with how I would rank them: Mata, Hazard, Oscar.

At the back, Cahill’s goal was a nice bonus and with Terry out for four more games, he at least deserves a cursory glance. With the next three home games against Man Utd, Liverpool and Man City though, along with tricky trips to Swansea and West Brom. I don't see a ton of value in Cahill in Terry's absence.

Everton
Howard, Coleman, Distin, Jagielka, Baines, Mirallas, Neville, Osman, Pienaar, Jelavic, Anichebe
Subs: Mucha, Heitinga, Oviedo, Naismith, Hitzlsperger, Gueye, Duffy

Not a whole lot to report here either. Fellaini missed out with injury, as we already knew, and may or may not miss another game this week. That meant Anichebe got another game but he isn't viable given the depth of talent elsewhere in this team. Pienaar will miss the Merseyside derby but with the suspension being for just a single game I'd be okay holding onto him.

Fulham
Schwarzer, Riether, Hughes, Hangeland, Riise, Sidwell, Baird, Richardson, Rodallega, Berbatov, Petric
Subs: Stockdale, Senderos, Kasami, Karagounis, Diarra, Dejagah, Kacaniklic

Kacaniklic has been dropped a couple of times now, and while he's likely only your second on third, it's time to start looking elsewhere as you can't be carrying dead weight as the inevitable suspensions and injuries start to rack up. The inclusion of Rodallega (#20), Petric (#10) and Berbatov (#9) was a touch surprising and based on the average positions, it was actually Richardson (#15) who took up the most advanced position. I'm not reading too much into this yet, but his odd deployment bears mentioning given the almost cult status Berbatov enjoys among fantasy managers.

Liverpool
Jones, Wisdom, Skrtel, Agger, Johnson, Gerrard, Allen, Sahin, Fernandez Saez, Suarez, Sterling
Subs: Gulacsi, Jose Enrique, Assaidi, Henderson, Downing, Carragher, Shelvey

Liverpool were unchanged this week which isn't overly shocking looking at their uninspiring bench, though the fact that Wisdom was able to hold off Jose Enrique on his return is notable. Honestly it seems as if Enrique (and perhaps Downing) have been highlighted as the players who represent the past and are being a little scapegoated by current management. I don't watch enough Liverpool games to comment with certainty, but it seems unlikely that the parade of youngsters earning time in this side are all better than some of the older options. Either way, the value in this side rests with the Suarez and Sterling, with Gerrard perhaps representing a risky play for those looking for options in their midfield.

Man City
Hart, Clichy, Lescott, Kompany, Richards, Milner, Barry, Nasri, Toure, Balotelli, Tevez
Subs: Pantilimon, Zabaleta, Dzeko, Sinclair, Kolarov, Aguero, Toure

The fact that Tevez and Aguero both didn't start despite Silva's absence is a real concern and I'm almost at a point where I simply abandon this City side, other than maybe Yaya, though his value has taken a knock of late too. With Dzeko getting among the goals again this week he too is sure to get some starting minutes in the future, making this front four a total mess. At the back, Richards seems to have found some favour over Zabaleta though that too is likely on thin ice given Mancini's past.

Man Utd
De Gea, Da Silva, Ferdinand, Evans, Evra, Valencia, Scholes, Carrick, Welbeck, van Persie, Rooney
Subs: Lindegaard, Anderson, Giggs, Hernandez, Nani, Powell, Wootton

De Gea keeping his place between the sticks is potentially noteworthy, though with another shaky performance at the back, I'm not sure we're too bothered either way. The situation up front has resolved itself with van Persie and Rooney showing an ability to play well together (at least by way of fantasy returns) but the midfield remains a minefield and should probably be avoided right now. For long stretches last season I had six players from the two Manchester teams, now it's very conceivable you'd be best served with zero.

Newcastle
Krul, Simpson, Williamson, Coloccini, Santon, Gutierrez, Cabaye, Tiote, Ben Arfa, Ba, Shola Ameobi
Subs: Harper, Anita, Cisse, Perch, Obertan, S Taylor, Ferguson

Given his well publicised success against Sunderland, it's very likely that Ameobi got the start here based on a handful on previous encounters with Newcastle's local rivals, which I can't say I'm a fan of. Either way, I don't see that as a long term trend and would expect Cisse to return soon. The same can be said for Steven Taylor, who presumably didn't quite pass the fitness test to return here. This side continues to offer good fantasy options at a number of positions, though Tiote's absence will have some impact on their usually decent defense.

Norwich
Ruddy, R Martin, Turner, Bassong, Garrido, E Bennett, Johnson, Hoolahan, Tettey, Pilkington, Holt
Subs: Rudd, Snodgrass, Howson, Jackson, Surman, Morison, R Bennett

Paul Lambert's former team is a lot like his new one, with players rotated ad nauseum, often including those with the most talent. I understand a manager's job involves more than simply playing the best players but the rotation of Snodgrass and sometimes Holt leaves me baffled to say the least. Nevertheless, they got the win this week against a good Arsenal side, and while I might argue with ESPN's notion that they deserved it given their 28% possession and 3 shots on target, it's a good 3 points for Hughton's side. Fantasy wise though, I see nothing here other than perhaps Holt.

QPR
Julio Cesar, Bosingwa, Diakite, Nelsen, Traore, Mbia, Granero, Park, Taarabt, Hoilett, Zamora
Subs: Green, Ferdinand, Cisse, Wright-Phillips, Mackie, Onuoha, Faurlin

We're starting to see a first team emerge here which could make the likes of Granero, Hoilett and even the mercurial Taarabt worth looking at once again. I'm not personally there yet, but one change in the front six this week is a step in the right direction and with home games to come against Reading, Southampton, Villa and Fulham up next, Hughes' men might finally offer some potential as reliable squad men.

Reading
McCarthy, Cummings, Gorkss, Mariappa, Shorey, Kebe, Guthrie, Leigertwood, Karacan, McAnuff, Pogrebnyak
Subs: S Taylor, Gunter, Pearce, Le Fondre, McCleary, Robson-Kanu, Roberts

The core of this side looks pretty settled these days but with the second worst attack in terms of shots inside the box and the league's fourth worst defense by the same metric, it's still tough to see a ton of value here. I'm not enamoured by his upside but Pogrebnyak at least seems to have guaranteed minutes, which shouldn't be ignored given his price tag and thus he deserves mention. Gurthie and McAnuff also place fairly well among their peers, though once again their chief appeal is their low price.

Southampton
Boruc, Clyne, Hooiveld, Fonte, Yoshida, Puncheon, Schneiderlin, S Davis, Lallana, Rodriguez, Do Prado
Subs: K Davis, Lambert, Ward-Prowse, Mayuka, Seaborne, Chaplow, Reeves

Boruc came into the team in goal, making him the Saints' third keeper for the year. I can't comment with authority on the finer points of shot stopping or when to come out for crosses having routinely shunned 'keeper duty since I was about nine years old, however, I do know that if you allow 92 shots inside the box and 137 in total in 8 games, you're going to concede goals even with Buffon between the pipes. We're not here to talk about this defense though.

The headline this week was Lambert's benching  and while you don't like to overreact to one game, this is the second time he's been benched. The good news is that those two benchings have come in tricky away games so his value isn't massively compromised, though it's enough to make the likes of Berbatov, Kone or Holt a touch more attractive. We'll touch on him more in the fanning the flames piece but I must note (read: thank) Lallana for his goal this week, which bailed my team out of another otherwise miserable effort.

Stoke
Begovic, Cameron, Huth, Shawcross, Wilson, Nzonzi, Whitehead, Walters, Adam, Kightly, Crouch
Subs: Sorensen, Palacios, Jones, Owen, Upson, Etherington, Wilkinson

Nothing really to note here as once again Stoke provide a consistent lineup. Though their generally solid lineup gave up four goals against United, it was actually encouraging to see them score a couple with Kightly quickly joining a solid list of mid-range midfielders.

Sunderland
Mignolet, Gardner, O'Shea, Cuellar, Rose, Johnson, Larsson, Colback, McClean, Sessegnon, Fletcher
Subs: Westwood, Bardsley, Campbell, Kilgallon, Vaughan, Meyler, Saha

Another settled lineup here with though my latest version of the model suggests some concerns over the sustainability of their success this year, but we'll get to that later in the week. On a personnel level there are plenty of good options here though so for now let's leave it there.

Swansea
Vorm, Rangel, Chico, Williams, Davies, Routledge, de Guzman, Ki, Britton, Hernandez, Michu
Subs: Tremmel, Graham, Dyer, Monk, Shechter, Moore, Tiendalli

I'm pleased to see Hernandez break into the first team though it's a shame that came at the expense of Dyer who has been a useful fantasy option over the past year and a bit. Michu played the role of nominal striker, though in truth he didn't play much further upfield than in past weeks. Nevertheless, he's a good option for his price. I'm not out on Graham and Dyer just yet but it looks like we might need to part ways soon.

Tottenham
Friedel, Vertonghen, Gallas, Caulker, Walker, Sandro, Huddlestone, Dempsey, Sigurdsson, Lenon, Defoe
Subs: Lloris, Naughton, Dawson, Falque, Livermore, Townsend, Adebayor

Vertonghen appears to have settled into that left back quite nicely which is a slight disappointment as I thought Naughton might develop into a decent fantasy prospect this year. Regardless, Gallas and Caulker both come at a discount price and my previous concerns about the latter have been dismissed with some decent outings which seem to have confirmed his place in the first eleven. With Walker and Vertonghen coming at a huge premium and the uncertainty between the sticks, it would appear that Caulker might bet he best bet here.

The midfield still isn’t really sorted as while Dempsey and Sigurdsson played together, one will likely find themselves on the bench again as soon as Bale returns to action. I’m still finding myself turned off by this whole attacking unit other than the Welshman. Defoe is worth holding if you own him and though I’m not confident in this assessment, I still believe he and Adebayor will split time in the coming weeks, making both tough to own.

West Brom
Foster, Ridgewell, Olsson, McAuley, Tamas, Yacob, Mulumbu, Dorrans, Morrison, Fortune, Long
Subs: L Daniels, Popov, Rosenberg, Jara Reyes, Lukaku, Gera, Odemwingie

I continue to be frustrated by Odemwingie and Long not getting enough playing, but Clarke is doing a decent job by all accounts and Long has three goals to his name, so what do I know (if I was pushing an agenda though, I might point to his four shots on target for the year to those now labelling Long as a 'natural finisher'). Regardless of the situation up front, Morrison remain a useful option and this defense should have better days again in the future.

West Ham
Jaaskelainen, Tomkins, Collins, Reid, McCartney, Noble, Diame, Benayoun, Nolan, Jarvis, Carroll
Subs: Spiegel, Cole, Maiga, Spence, O'Neil, Chambers, Hall

There's a lot to like about this side, who offer consistent lineups as well as upside at a number of positions. We'll touch on Nolan and Noble in this week's fanning the flames and there's not much more to say for now.

Wigan
Al Habsi, Caldwell, Figueroa, Ramis, McCarthy, McArthur, Boyce, Beausejour, Kone, Maloney, Di Santo
Subs: Pollitt, Jones, Watson, Gomez, McManaman, Boselli, Miyaichi

I waxed lyrical about this front three a couple of weeks back, and while they weren't able to contribute on the scoresheet this week, they showed flashes once again with Maloney in particularly looking most likely to nick a goal. With his goal this week (a beauty if he meant it by the way), there will be plenty of managers jumping on the Boyce bandwagon, and while I'll expand on this in the later piece, be careful about buying a defender from a side who just aren't very good at the back because he might get the odd goal or assist here and there. 

Saturday, October 20, 2012

Gameweek 8 Preview 2.0

I am away this weekend so I'm not going to have time to write up how the below numbers have been generated, but it's essentially the format prescribed in the recent goals and assist pieces, only limited to teams for now. I'm posting the data now so we will have a real test set of data from this weekend to work with, which will hopefully highlight a couple of areas where the model could be improved (and by that I mean where the process, rather than the result, is wrong. If Stoke hold United to no goals on 27 shots, the model isn't necessarily broken, it was just a bad day at the office for van Persie and co).

xG Shots - expected goals scored based on the shot data they have registered and the shots given up by their opponents, regressed to a team average conversion rate (not a league average).

xG Chances - expected goals scored based on the chances created by a given team and the chances surrendered by the opposition, regressed on a team average conversion rate (not a league average).

Finally, apologies for the lack of reader questions this week - normal service will be resumed for GW9 so get them in early on Twitter, Facebook.