Wednesday, December 18, 2013

Gameweek 17 Preview

Tuesday, December 17, 2013

The Aguero Problem

Given the number of transfers already made (290k at the time of writing), this post is potentially too late, but for those Aguero owners who haven't yet pulled the trigger on a replacement, here are a few factors to consider before you do.

The Argentinian's absence seems to be being pegged anywhere between four and six weeks, which means he will almost certainly miss gameweeks 17-21 and his absence could potentially extend through GW22 or even 23 (with good squad depth City should be mindful not to push their talisman and risk further injury as seems to have been the case with Kompany in the past). With his return therefore scheduled to coincide with the middle of the second wildcard window, the cost to buy Aguero back should essentially be ignored and value for money for right now is a factor. This contradicts the situation if he was missing for just a week or two, in which case it might be worth eating his cost to sit on the bench rather than lose 0.5m when you are forced to buy him back in two weeks.

So if we accept that value is a consideration, the assumption that Rooney is the best (or only if you read some sources) option looks arguable (that's assuming, of course, you already own Suarez. If you don't, you should probably just stop reading and go grab the Liverpool front man).

Team Forecast
If we look at the team forecast for the next six weeks, Aguero's absence become an even bigger blow as the model loves City compared with everyone else in the league.

One can argue, perhaps, that their fortunes will take a hit without Aguero, though in Negredo (and possibly Dzeko) they have a ready made replacement so it's reasonable to conclude that they'd still place atop these rankings even if we could eliminate Aguero's impact. Though Rooney may personally have being playing better of late, the model is terrified of United's prospects and is pretty suspicious of a turnaround, aside from the promising returns for the next three easier games. Based on their team's prospects (which are probably more reliable than individual forecasts as they're predicated on larger shot samples), Negredo looks like the best play here, followed by Suarez, Giroud and then Rooney or Lukaku.

Individual Forecast
Ok, so if you don't have Luis Suarez, seriously, stop reading and go and buy him . . . now. It's okay, we'll wait for you. Ready? Right, assuming you already have Suarez locked down, and most likely paired with someone else on this list, what now? Though not a necessity, it's always nice when the model backs up your own instinct based on watching games, and that's the case here as the next four options are those who likely sprung to everyone's mind when Aguero went down. The model doesn't see a great deal between this foursome (Giroud, Lukaku, Negredo and Rooney) and the only comment I'll add here is that if Giroud is the man you target, you may as well go with someone else for this week given his tricky fixture (CHE) and then bring him in for GW18. Also, one small factor is the captain's armband, and we can see that based purely on the data above, Lukaku is the only man who would be captained twice (GW18 SUN and GW21 NOR) over Suarez. Not a major point, but a point to consider in a close contest like this.

As we said above, given Aguero's expected length of absence, cost should be a factor. Of the four names noted above, Rooney costs between a 1.9-2.7m premium; money that could be used elsewhere to take advantage of other promising players. Those mouth watering numbers for City shown above coincide nicely with Silva's return to the side, for example, and he should be at the forefront of your plans as a high upside differentiator. Pairing the Spaniard with his teammate Negredo would given you an average expected points haul of 13.5 points per week between them, a total that simply cannot be matched by pairing Rooney with anyone available for 1.9m less than Silva (Ramsey would be the best bet at 1.6m less with an 11 point per week average). Granted, that Silva forecast is based on a small sample so I wouldn't necessarily want to hang my hat on it, but the point is, 1.9m-2.7m is a serious premium that can be used elsewhere in your team.

Suarez 46%
Lukaku 32%
Giroud 28%
Rooney 19%
Remy 15%
Benteke 10%
Lambert 10%
Negredo 9%

The latter three names enjoy an advantage here, though only Negredo has the data and upside to really come close to filling Aguero's shoes. The gap to Rooney isn't huge, and some of the Englishman's owners will be dead teams who picked him months ago, though it still represents an advantage that is becoming increasingly hard to find (that of being a "differentiator" if you can call a 9% owned player that).

This isn't an easy question though in some ways it's a welcome one, as while Aguero was playing out of his mind, virtually every contending team owned him so all he really did was make everyone compete with ~92m or so and inflate our collective weekly scores. With Rooney and Negredo looking like the favoured replacements at least we're seeing some element of distinguishing between our teams. As for what the ultimate answer is, I'd suggest that the one based in 2013-14 data is Lukaku, or Negredo if you want an element of differentiation (which I would personally endorse). Could Rooney outperform them both? Sure. But that proposition is based on elements that this blog doesn't really deal with, like achievements from prior years or the ever mythical "form". The way I see it, you have a player who's fantasy points have increased thanks to five goals in the last six weeks, but also one who's production looks unsustainable based on his shot totals over that period (based on his 23 shots, we'd expect 9 to hit the target and 3 to hit the net).

Negredo's data isn't spectacular either, and he too has benefited from a favourable G/SoT rate (6 goals / 11 SoT), but he also finds himself in a team whose attacking fortunes look much brighter, comes at a 1.9m discount and offers an increasingly valuable differentiation factor. Given that it looks like everyone will be captaining Suarez with a regularity harping back to the days of Ronaldo, I'd suggest you can take a risk with someone like Negredo over Rooney, especially if that could open the door for you to also bring in someone like Silva. In fact, that final factor may be the deciding piece of the puzzle and could finally offer an alternative to those wanting to go another way than Ramsey whose own 62% ownership has become an issue.

If in a position of strength, one can also make a good case for Lukaku who falls in the middle of most of the above considerations, though his high ownership rate suggests that many of you (including your humble blogger) already have the Everton man locked down. If that doesn't apply to you then the Belgian can offer increased stability and Premier League pedigree over Negredo, but comes with a lower ceiling and doesn't do much to really set you team apart.

Friday, December 13, 2013

Going streaking

Whether you're lucky enough to still have your original wildcard stashed away in your back pocket or not, as we approach the second wildcard window, it's worthwhile putting together a transfer plan to best exploit the fixtures on offer. We have at least four transfers before then (GW16-19) followed by the important period right after we play the wildcard, so with that in mind, let's try and identify a few fixture streaks during which we might want to target specific players.

For simplicity I've only looked at streaks of four gameweeks here. In reality we might want to dig deeper into streaks where you like three out of four games, or four out of six, but this is the starting point, for better or worse. Double gameweeks will also likely cause chaos at some point too, but without knowing where or when they'll strike, their impact is ignored for now.

The below table shows the expected goal total over the four highlighted games above or (below) the team average. So a score of -1.5 suggests that the model believes Arsenal will score 1.5 less goals over those four games than their average GPG to date would suggest (that GPG is based on actual goals scored/conceded, not regressed or expected goals).

I'm going to keep the analysis brief as this data really speaks for itself, but there are a few comments to make on a couple of these streaks.

Arsenal and Liverpool have been firm favourites of fantasy managers to date, yet appear to be risky investments in the immediate future based on the above (more on Arsenal in the longer term below). Both face three away games in the next four, including trips to the likes of the Etihad, Stamford Bridge and White Hart Lane. Does this mean you should be targeting Suarez and Ramsey for sale? Probably not, as they still do very well in the weekly rankings, but the lesser players from those sides should be considered expendable, particularly on the defensive end of the pitch.

Sunderland look like they enjoy incredibly easy fixtures as they're forecast to well exceed their goals scored and conceded tallies to date. However, this is more a product of the model thinking they've been somewhat unfortunate to date rather than the fixtures being unbelievably favourable. Saying that, over the next eight gameweeks they do get a number of promising fixtures (NOR, @CAR, AVL, @FUL, SOU, STK) so players such as Giaccherini or Johnson could make useful 4th midfielder types in paired with the right complementary piece.

The model loves Arsenal after GW19, and well it might given their fixture list of @NEW, CAR, @AVL, FUL, @SOU and CRY. This is one situation when careful planning is required, as to potentially get a 1.5 goal bonus on top of the excellent form this team has already shown suggests we should really be trying to squeeze an extra Gunner in our team for this period (the start of that period nicely coincides with the opening of the transfer window so they should be a hot commodity for wildcarding teams).

Another team whose assets you will want to maximise coming out of the transfer window is Man City. They get CRY, @SWA, @NEW and CAR from GW19 - 22, a period of fixtures that suggests we should try and squeeze at least one more attacking player into our sides alongside the presumably ever present Aguero. It isn't always easy to identify such players in this changeable City side, so here's hoping we're seeing some vintage David Silva play by the time Santa delivers that appetising run of games.

West Brom are a massively underrated defensive side with ownership numbers of 3.0% or less across the whole back five. Granted, the results of late have taken a turn for the worse (no clean sheets in the last five games) but they still boast excellent +/- scores of -19% at home and away and have four teams on deck who are all struggling for goals right now (@CAR, HUL, @TOT, @WHU).

Saturday, December 7, 2013

The Second Act

As we all know, the fantasy season does not comprise 38 independent events. Medium and even long term plans need to be put in place based on the next 'x' number of games and over-reliance on the now can handcuff your team in the future. Though all teams are constructed differently and thus will encounter these periods at different times, there are also overarching factors which impact all (or at least most) teams in the same way.

Sometimes these are obvious, such as the opening of the second-wildcard transfer window or the real-life transfer window which potential opens the door to new options for our teams. On other occasions, however, the trends are tougher to spot, especially when we're in the eye of them (they are much easier to see with hindsight). We are potentially at a changing point for one of these period-changes.

Like any mostly free market, the demand for players will settle into something of an equilibrium, and for most managers the leading role in determining their level of demand is past performance (yes, if you're reading this you'll almost certainly place more emphasis on future forecasts, but we'd still be hesitant to back a player with a terrible points tally to date, no matter what his underlying data suggested).

Consider the below chart:

Ignoring defenders and 'keepers for a second (whose ownership numbers tend to be a bit more unpredictable), we see just eight players who've amassed 50 or more points and are still owned by less than 10% of managers. 16 such players are owned by 10%+ with half of those owned by a full 20% or more. This is always going to be the case to a degree (the best players, or at least those with the best results, attract the most managers), but it seems as though we're reaching a critical mass with regards to this trend. 

Just look at this past gameweek. Nine midfielders and forwards managed to notch double figures but four of them seemed to be owned by everyone (Suarez, Yaya, Ozil and Hazard) and four were - justifiably - owned by no one (Oviedo, Delph, Agbonlahor and Dyer) with Lampard being the sole occupier of the middle ground. Throw in decent contributions from other favourites like Aguero, Ramsey and Lukaku and you're left with a situation where unless you owned all four star performers, a couple of the middle guys and notched a couple of clean sheets, you probably ended up losing ground in the overall leaderboard this week (my own team lost over 4,000 places despite what appeared on its face to be a very healthy 88 point week). 
This all conspires to a very long winded way of saying it's time to push for differentiation.

Now, before we delve into names, let's just lay down a couple of ground rules on what we really mean by this. First, differentiation just for the sake of it should not be pursued. Sometimes there are good reasons for people to be suspicious of a given player and a need to 'stand out' shouldn't interfere with the ultimate goal of points maximisation. Second, even at this relatively early stage of the season, different teams will have different appetite for differentiation based on their league position and objectives. If you've amassed a 100 point lead in your mini league, and your ultimate goal is to win that, rather than strive for the highest overall points finish, then picking Yaya Toure - who is owned by your next three closest competitors - makes more sense than if you're in the chasing pack. I would personally caution against such 'defensive' moves this early in the season, but that's something for each manager to decide.

With the customary caveats in place, let's have a look at some of the most widely owned players right now and see if we can find some reasonable alternatives. For the sake of repetition avoidance, I'll add this disclaimer here just once. These are ideas designed to give some fresh thought to what is becoming an increasingly stagnant fantasy season. I wouldn't (indeed won't ) make the majority of them and I fully understand that they will all likely involve a lot more risk than sticking with who we "know" will have future success (this in itself is a misnomer, but that's a point for another day.

Aaron Ramsey (62% owned)
The Welshman's 62% ownership number is a huge outlier in the earlier chart, though in all honesty, this number could really have peaked higher given the way he's played to date (the 'real' ownership number is much higher, of course, as that 62% is based on a denominator which includes a lot of teams who have gone untouched since GW1 and still roll out the likes of Podolski (1%), Adebayor (0.7%) and Zaha (1.3%)). Ramsey simply leads all midfielders in total points despite opening the year with a price tag of 3.0m+ less than all his closest rivals and appears to have turned the corner from 'someday prospect' to 'everyday star'. 
The model supports Ramsey's high ownership to a degree as his 61 xP ranks 3rd among midfielders (behind the surprising Gerrard and Hazard) but it also suggests that his future forecast is closer to players available for comparable prices and thus he isn't an automatic pickup. Let's filter our data to highlight a couple of options. We're looking for a player who:
  • is available for close to Ramsey's price (we'll give ourselves a 1.0m variance, so a maximum of 8.3m),
  • gets a sufficient number of chances per game (more than 1.0 SiB per 90 minutes),
  • does something with those chances (more than 0.75 SoT per 90 minutes) without benefiting from unsustainable good fortune (SoT% of 50% or less), and
  • can also contribute in the assist department (more than 1.0 CC per 90 minutes)

There's some intrigue here for sure, with Paulinho's 1.9 SiB90 and Sigurdsson's 1.4 SoT90 jumping off the page. However, the first Spurs man seems to be taking too many ill advised shots given his 32% SoT% rate while his Icelandic teammate is hitting the target with long range efforts, but as we know, these tend to hit the back of the net with less regularity than those from close range. Mirallas is probably the safest play here given his solid returns and supporting data yet his tendency to be withdrawn early in games and a lack of explosive upside makes it really tough to part ways with Ramsey for the Belgian wide man right now. As keen as I am to suggest alternatives to Ramsey and as likely it is that he regresses in the coming weeks, no obvious one-for-one replacement is apparent. As we move into wildcard season though, alternative arrangements should certainly be considered.

Yaya Toure (38%)
Here we're looking for reliability. Yaya has about as high of a floor as you can find and can contribute solid value over virtually any stretch of games. To replace him, we're therefore looking for the below attributes:
  • ability to score with some consistency (more than 0.75 SiB)
  • maximises his opportunities (more than 0.75 SoT)
  • strong chance at assists (more than 1.1 CC)
  • plays consistently (more than 900 minutes played or return from injury will guarantee playing time)
  • less than 10% ownership (if we're sacrificing his consistency we need to make gains on a lot of other managers)

Of all the alternatives looked at today, this is probably the one that makes the most sense. Some of the best options here currently carry injury concerns so may not considered immediately (Walcott, Michu and Silva) but Coutinho looks well healed and took up some very promising positions in Sturridge's absence, behind the all conquering Suarez these past couple of weeks. Though it's a tiny sample size and too much weight shouldn't be placed on this data set, the fact that Walcott has tripled Toure's SiB90, doubled his SoT90 and doubled his CC90 suggests he could offer significant upside when healthy and should be monitored closely for signs that he's back in Wenger's first team plans. Finally, I'd be remiss if I didn't draw some attention to Frank Lampard who in many ways looks exactly like Yaya Toure, just with a lower G/SoT rate. Lampard's currently sits at 33% versus a 42% career rate while the Ivorian's is hovering around 64% compared to a career rate of 30%. If it's reliability you desire, it seems that Lampard should be able to fill a similar role for 2.0m less and with a quarter as many managers sharing in your spoils.

Luis Suarez (39%) / Sergio Aguero (33%)
We're not looking for much here. We just need someone who is averaging over three shots a game, with two on target and . . . ok we can stop now because we've already filtered the entire league. Only five other regular starters (Lukaku, Sturridge, van Persie, Grioud and Remy) are able to average as many SiB as this pair have averaged SoT and none of that group come close to boasting the assist potential of this elite duo. I hate the label "must own" and for the most part this has been wrongly attributed to various players over the years but right now the only real argument against owning this duo is one of value as you could go with a Lukaku-Giroud-Remy trio and save at least 4.0m, but then it isn't really clear who you'd then spend that surplus on. The return from injury of Benteke, Sturridge and van Persie will at least swell the pool of players who could threaten this pair's dominance but of of now, the only reason to look elsewhere would be injury to either of these forwards.

Olivier Giroud (34%)
Given what we've said about Suarez and Aguero above, Giroud is essentially competing with every other forward in the league for the right to be your third front man. Though Giroud does offer some assist potential, we can't be too choosy as we slide down the price list, so to fill this role we're merely looking for a player who:
  • offers a consistent goal threat (over two SiB per 90 minutes),
  • costs the same or less than Giroud (with Suarez and Aguero we're already in danger of committing too much cash to our front men with Giroud),
  • is owned by less than 12% managers (that cut off is clearly fixed so I can include Benteke
  • offers some threat of assists (more than 1.0 CC per 90 minutes)

Giroud has enjoyed a great season to date but all of the above options either strong differentiation potential, cost savings, or both, and given Arsenal's fixtures in the immediate future, this is one move that could be considered for sooner rather than later.

Ross Barkley (28%)

I would imagine a fair number of Barkley's owners still hold him due to more pressing issues in their teams eating up weekly transfers, but still, he's held by an awful lot of managers and with wildcard season approaching, we should be considering everyone's place in our respective squads. There are several options to really like here with Arnautovic particularly catching the eye. The Stoke man has taken a very healthy share of shots given his price bracket, even if the number of target somewhat disappoints. However, his 20% SoT% rate is well below the 33% rate he posted in the Bundesliga while at Bremen and so if that number starts to track back towards his expected average, it should give him a profile to far exceed the somewhat mediocre prospects of the majority of his peers. Digging deeper we have Redmond who is enjoying his prolonged run in the Norwich side following their rash of injuries, and even deeper we have the highly rated Jonny Williams who has finally returned for Palace and could offer intriguing value at just 4.3m. In short, this is a category with a good number of alternatives and if you've transfers to spare, Barkley's waning fortunes should really be taken notice of.

Thursday, November 14, 2013

Getting defensive

Attack-minded defenders have long been considered one of the key ingredients to a successful fantasy team, and while I may personally suggest this quest is sometimes overpursued, there's no arguing with the overall premise that such a player can be extremely valuable.

Two issues arise though.

First, managers tend to overvalue this threat, paying huge premiums which simply cannot be justified. The most obvious example of this in recent times has been Leighton Baines (who has rarely justified the huge premium required to get him) though we've seen similar trends in the past with the likes of Thomas Vermaelen, Nemandja Vidic and even Joleon Lescott.

Second, managers tend to overvalue past success, which can be ill advised at the best of times but is particularly dangerous with defenders. With a few notable exceptions, defenders tend to enjoy limited chances to earn points and the sample sizes involved are generally very small. Thus, a player with the odd goal coming from a handful of shots is no more likely to score again than any number of comparative players who haven't yet registered a goal this season. Of course, we're not saying those "unlucky" few are any more likely either, but the latter are almost certainly held by fewer managers and probably come with a lower price and thus deserve our attention.

The chart below can be sorted by team and plots a player's price against his expected points purely from attacking means (xP). We're trying to identify players whose production exceeds or matches those available in his price bracket and this is best done on a team by team basis (as otherwise you are trying to compare too many variables at once). Once you identify the 'best' options from each team, you can then move on to comparing them all considering other factors.

A note on expected points. In the model, two bonus points are awarded for each goal, based on a approximate study of how these are generally handed out. Hence, if a player has racked up enough shots to score exactly one goal, his xP would be eight rather than the six you might expect.

Sagna and Mertesacker have produced the same attacking points to date with the Frenchman ahead by a couple of points in xP. Given this fact alongside Mertesacker's much higher ownership number (15% vs 1%) and Sagna's slight discount, Sagna looks like the better pick if you were choosing one today, though Mertesacker owners shouldn't feel overly rushed to move him.

Aston Villa
There's really nothing to see here and if you want to invest in this back line you should do so based on playing time and price tag. That could lead one to settle on Baker given his 3.9m price, assuming you are confident on him holding off the challenge of Clark when Luna resumes his left back duties. This isn't a great defense but they've been reasonably good both at home (-1% SiB +/-) and away (-3%), but don't expect to get much on the attacking side of the pitch from your Villains.

Caulker is the only defensive player to notch any points, and coupling that with his name recognition leads to a potentially dangerous situation. Four total attempts, with two on target strongly suggests that his two goal haul to date has been fortuitous and thus that 0.9m premium over Ben Turner looks unjustified.

Ivanovic leads all defenders in xP to date with a very impressive 17 and managers have done well to recognise this, with a 25% ownership already racked up. With a risk rating of 1.2, he looks as safe as anyone in this team and the other 25% of managers who currently own Terry, or particularly Cole and Luiz, should strongly consider making a move to Ivanovic next time they have a spare transfer (or use their wildcard).

Crystal Palace
Delaney owns a slight advantage in xP here, but not enough to justify his higher price tag. If you are interested in this defense (hot tip: you shouldn't be) then Moxey - who boasts a perfect risk rating of 1.0 - is your man.

To date, this one is a no brainer. Coleman trails Baines in actual points by five but that total is far from enough to justify the extra 2.2m required outlay. Add in the fact that Coleman actually tops Baines in xP and you see that the young Irishman should be the pick here. One caveat which the model doesn't account for is Baines' threat from the spot but it's highly unlikely that Everton will earn enough spot kicks for Baines to deliver superior value to his cheaper colleague. The 17% of owners still holding Jagielka - who offers a reasonable though unexciting attacking threat - should consider a move to Coleman when re-shuffling their pack.

You really don't want to touch this entire team right now and there's nothing here to change that well reasoned opinion. Riether enjoys a marginal advantage in the attacking threat column but is well overpriced at 5.0m, even if he does carry a low risk rating. Just don't bother, please.

Davies boasts by far the best xP and actual points to date but seeing as you're surely only employing a Hull defender as a minimum price bench filler, 4.6m seems too steep a price to pay for only a marginal attacking advantage. This isn't a bad defense (SiB +/- of -2% at home and away) and they certainly hold value as 5th defender types but I'd personally lean towards one of the 4.0m options with Rosenoir looking nicely settled in this lineup at right back with Elmohamady moving further upfield.

Glen Johnson has been the best attacking threat to date, yet hasn't been rewarded for his efforts so far. Purely on this added threat, he doesn't appear to justify the extra cost, but given his added security in the first team (he's yet to be benched and his risk rating is only impacted by injuries) it might just be worth looking at Johnson if you either don't have Mignolet or want to double down on this useful unit. If Cissokho really locks down that left back spot it becomes arguable that Johnson's superiority in xP is insufficient to justify the extra 0.7m outlay but for now the Englishman just about holds the edge.

Man City
Not unexpectedly, Kolarov leads this team in xP yet - again, as expected - it's his risk factor (2.8) which renders him very hard to own unless your team budget has swelled to the point when you can routinely sit a 5.3m player. Even then it would be hard to own him as you're never quite sure who will start but if you feel you have a good read on Pellegrini's team selections, it's hard to argue with Kolarov's upside. Elsewhere no one really excels and you're faced with a contest between Nastasic and Zabaleta who are closely matched in terms of risk factor and attacking points potential.

Man Utd
Purely on attacking threat, Evra is just about worth the extra expense over his cheaper colleagues, but when you factor in his risk rating, he becomes even more attractive. Jones' occasional deployment in midfield has been something of a non-issue to date with the utility man failing to offer much going forward even when deployed further upfield. At 0.5m cheaper than Evra he's a tempting proposition though with Rafael on the mend to vulture a large slice of the right back minutes, the ever reliable Frenchman remains a solid investment.

Putting this week's poorly timed suspension aside (the Magpies get Norwich at home next), Debuchy is the clear pick here and is one of the more extreme outliers in this piece. He enjoys a perfect risk rating, leads the team in actual attacking points which are totally supported by his xP, yet he's owned by just 1% of owners, only good for 3rd on his own team. This isn't a great defense so if you're planning on benching a Newcastle player most weeks then perhaps Yanga-Mbiwa at 4.4m makes some sense, but if you plan on starting him, Debuchy's high floor and ceiling make him a useful pickup.

No one really stands out here and thus it seems prudent to defer to risk rating rather than attacking potential, which would point us towards Turner as the best investment here. Whittaker's owners - presumably not paying attention or putting out other fires - should obviously look to divest with their man offering neither security nor upside.

Lovren's 11 attacking points are nice, but his price is getting dangerously high and if you're looking to buy now, you're too late. Clyne should be a consideration given his significantly lower price tag and increasing security (he's played every game since GW4), ahead of Shaw who seems to face a more sustained threat from Fox. At 5.4m, Fonte might be the 'Goldilocks' option with a nice balance of security, price and upside, though if pushed I would personally lean towards the potential value of Clyne.

Huth and Shawcross just about justify their half million premium over Cameron and Pieters, though it's probably close enough that you wouldn't rush to make a move if you owned either of the full backs. Fans of differentiation will want to note Huth's 2% ownership versus Shawcross' 10%.

Bardsley's heroics this past weekend catapult him atop these rankings, yet his risk rating suggests he's all but impossible to buy into. Sunderland have been a very solid defensive unit at home and there's room to add a Mackem as a 4th / 5th rotation type defender, yet it's hard to really pinpoint anyone who offers too much excitement. Looking purely at this team, O'Shea's low risk profile looks appealing, but higher upside can be found on other teams in his price bracket and thus if you want to buy in, you might need to take a calculated risk that Brown or Bardsley can lock down their places at a reduced cost.

Chico's position on this chart will surprise many (include your humble blogger) as Davies and Rangel often impress going forward for Laudrup's team. This isn't the solid unit we saw for much of the last couple of seasons, but at 5.0m Chico's attacking upside, playing time and at least some promise of defensive value make him an intriguing option.

This one isn't so much obvious who we should pick as much as who we shouldn't. At 7.2m Vertonghen is valued as a defensive player who contributes big time attacking scores on a regular basis yet he's not managed to achieve that in either the actual (3) or xP (5) columns. He's essentially offered the same profile as Dawson at a 1.8m premium which is essentially the same as choosing a 5.4m scrub over Rickie Lambert.

West Brom
With Jones' injury McAuley's overall profile now looks quite promising and he just about deserves consideration despite some discomfort in paying 5.4m for him. West Brom have a good defense though, especially at home (-16% SiB +/-) and McAuley offers a sufficient upgrade over his peers to make that 0.5m worthwhile.

West Ham
Reid leads the team in xP and ownership (6%) but it doesn't look like the former justifies the latter. We saw plenty of early turnover in this team which has limited the risk rating of some players, but with Rat and Tomkins playing the last six games, they can probably be considered to be reasonably safe in Allardyce's first team and thus their lower price tags make them more attractive compared to the comparatively pricey Reid. Collins - who has yet to be voluntarily benched - looks to be the best value of all, assuming he can put his injuries woes behind him for a sustained period, though be forewarned that you are signing up for very little attacking threat.

Thursday, November 7, 2013

Will some players always underachieve?

We had an interesting comment from a reader this week regarding the visualization posted plotting actual points with expected points. My proposition was that the players whose xP trailed their actual points by a distance were likely undervalued by the market, and while we wouldn't suggest they will somehow "make up" those points left on the table to date, we would expect their production to take an uptick assuming they continue to get chances and playing time at a relatively consistent rate. The reader had a different view:

"When I look at this chart I don't see underperformers or overperformers all I see is players on form who are capitalising on their chances (Ramsey and Rooney) and players who are of such quality that they will always out perform the normal (Aguero and Yaya) . . . I believe that if you reconstructed this table after xmas with a start date of tomorrow then the same players would occupy the two sides."

It's a fair proposition and one I wanted to examine further. I think there's a general discomfort with the idea of regressing players' production to the mean as it seems to suggest they are all created equally. A couple of responses to that:
  1. For conversion rates which appear to be repeatable year on year, such as shot on target percentage (SoT%), we regress players to their own historical rates (where available). This means that if we say Olivier Giroud has an unsustainable SoT%, we're not saying his is too high compared with Danny Graham or Frazier Campbell, we're saying it's way above his own historic rate.
  2. For conversion rates where we do regress to a league average (or at least use league average in a weighted average), it's because I haven't seen any evidence that players can consistently perform above the average in that given rate. The classic example is goals per shot on target (G/SoT) which tends to regress close to a mean for most players, with only a couple exceeding the average for more than a couple of years in a row (and that would be expected even if we were talking about a totally random event). There might be some repeatability there, but it's a lot less than most would expect based on purely on notions like "form", "class" or being "clinical".
The good news is that this is fairly easy to test. Below we've plotted players' +/- score as of this week (which shows the difference between their actual and expected points with a positive score meaning their expected exceeds their actual) against the same metric from the midway point of last season. I picked that point in time based on the reader comment about Christmas but I'm fairly confident a similar conclusion could be drawn from pretty much any two comparable samples:

The first observation is that we see very little correlation from year to year. 10 players outperformed their points total last season by at least 10 points, yet only one of these (Podolski) has managed to outperform his total to date by even 5 points. Similarly, eight players underperformed their underlying stats by 10 or more points last season, and of these two (Lambert and Cisse) have once again failed to match their live up to expectations. On the flip side we've seen players like Aguero, Rooney, Lallana, Suarez, Michu, Walters and Fellaini benefit or suffer from huge reversals in fortune over the two samples.

One of the things I love about sports writing is that it can be a gateway into so many interesting subjects, and while I'm not learned enough to talk about most of them here, I would venture that there is an element of bias regarding the way we judge the above. When a player like Ramsey explodes in a small sample, we tend to quickly absorb that information into our collective psyche and it becomes the new self evident truth that he is a great player (despite several seasons of reasonable yet unspectacular play, at least from a fantasy perspective). We then place too much weight on these recent events, much like how people stop swimming after a shark attack, despite the fact there are countless things more likely to really kill them that they ignore every day. I believe the term for this specific type of bias is referred to as the availability heuristic.

In the chart we see Aguero has the second highest +/- score for 2013 and one could rationalise that being due to his superior skill and quality teammates. Indeed, that's possibly true to a point. However, he had those very same skills and most of the teammates last year too, yet was actually one of the biggest underperformers last year, serving as a constant source of frustration for his owners. Or take van Persie. Last year he ascended to a new level and was casually thrown into conversations alongside the best in the world, and thus the fact he outperformed his xP by a full 13 points through half a year could be discounted as him simply being better than everyone else. Fast forward 10 months and we have a player who has only just caught up to his xP total for the year, having suffered through some bad luck these past couple of months.

As a final check, the colour coding relates to the players' team's league position ranging from 1st (green) to last (red). I wondered if we'd tend to see players from the better teams show an ability to repeat positive seasons as they benefit from more quality chances per game. I guess this works to a degree in that those in the bottom left quadrant generally play for better teams, yet there's not enough here to really draw any solid conclusions.

It's always good to challenge forecasts like the ones you find in these pages - especially the ones found in these pages! - but caution should also be exercised when dismissing data which contradicts our current view of the game. There are certainly aspects of a player's game which can consistently be above average (SoT% for one) but others seem far less repeatable. The current iteration of the model adjusts for these differences and thus that's why we're going to see turnover in the players who over or underachieve expectations. 

Points versus expected points

The below chart is something I've posted before and am tweaking for the new website so I thought I should post it now and get your feedback while also hopefully providing a quick view to identifying some players to research further:

Sunday, November 3, 2013

Gameweek 11 Preview

Post any questions in the comments below, on Twitter or on Facebook and I'll get to the best ones before this week's deadline.

Monday, October 28, 2013

Gameweek 10 Preview

The revised strategy for this season is to post the weekly preview data as soon as possible, giving you (a) the chance to use it to help with any early transfer decisions and (b) to collate questions on why a given player is so low or high, to be answered on Fridays before the transfer deadline.

Thursday, October 24, 2013

Gameweek 9 Preview

Given how early in the season we are, the model is still liable to throw up the odd outlier and so in these weekly posts I plan to address those, shall we say, unexpected results. In future weeks, the plan is to post the data as soon as possible after the final games' data is up and then you can raise questions/issues during the week, to be addressed on either the following Thursday or Friday. For this week, I'll just try and guess where the questions might lie:

Keiren Westwood
Sunderland have conceded at least two goals in six straight contests, yet the model thinks they'll do okay this week. What gives? Well, having conceded 7.3 shots inside the box at home, they're hardly a team without hope (that alone would be the 9th best total of the teams playing this week). Add to that the fact that Newcastle have averaged 30% less SiB against their opponents than average, while only averaging 6.0 SiB on their travels, and you get a game where we're expecting Sunderland to only concede a handful of good chances (5.3 SiB) which gives them their best shot at a clean sheet to date (36% based on historic averages for team surrendering those shot totals).

Seamus Coleman over Leighton Baines
In reality this ranking is too close to pick between and essentially the model is saying they are equal. Baines is actually worth a little more because we know he has a steadier source of shots from set pieces, but those from free kicks are of course built into the model already, yet Coleman still comes out on top. Coleman has accounted for 6% of Everton's SiB compared to just 2% for Baines, hitting the target more frequently (50% vs 25%) too. Baines has a very slight edge (15% vs 13%) in the created chance department but to date these players have been very close and the 2.1m premium looks tough to justify.

Mezut Ozil
As much as Ozil has impressed to date, his current forecast of close to eight points looks aggressive compared to his peers who top out at just six. The only real explanation for this is small sample size and some of his somewhat fortunate conversion rates which aren't fully regressed in the weekly forecasts. The main culprit is that 88% SoT% which inflates his shot expectation for the week, even when adjusted for his historic average. Saying that, the data suggests Arsenal to top 13 SiB and 20 total shots, which is almost unprecedented and with Ozil being a central part in everything good about the Gunners to date, it's tough to argue against him being the top pick this week, even if the margin is probably a bit smaller than the model currently suggests.

Where are all the Tottenham players? Paulinho (3.7) and Soldado (3.5) are their best options yet find themselves way down the rankings in 16th and 12th places respectively. With a home fixture against Hull, most probably expect them to murder their opponents this week, yet the data suggests otherwise. First, Spurs are only averaging 6.8 SiB at home with a +/- of just 8%, both of which put them in the same league as Sunderland and Norwich rather than Arsenal and Chelsea. Second, Hull have actually been relatively good as suppressing shots away from home and while the results haven't come, they can be forgiven for shipping goals at Chelsea, City, Newcastle and Everton. McGregor's ranking shows that the model doesn't think Hull can necessarily go to White Hart Lane and keep Spurs at bay, but a thumping is not the forecast result either which limits the upside of Soldado and company this week (though he and the other Spurs stars remain solid starters). 

Dousing the Fire, Fanning the Flames: Gameweek 8

As a quick introduction for those new to the blog, this piece runs every couple of weeks during the season and looks to shine a light on those "hot" players whose "form" looks unsustainable and those "cold" players who should enjoy success in the future if they keep playing the way they have to date. The below chart shows this week's subjects, with those on the left having outscored their underlying data and those on the right having outperformed their score.

Before we start, as is becoming a tendency on this blog, I need to add a quick caveat as to exactly what we're saying here. First, we are not saying that a player will somehow "get back" or "give back" their production to date or that bad luck will necessarily follow good luck. We are saying that players' (and teams') conversion rates should regress to the mean, seeing them earn points at a rate more in line with their underlying stats (which could be a good or bad thing, depending on which side of the below ledger you find yourself). Second, just because a player has outperformed their stats doesn't mean they don't represent a good pickup. If a 5.0m midfielder had 47 points to date but we believed he had only "earned" 36, he would place highly in the 'overperforming' rankings (-11) yet still represent the best budget midfielder in the league. Similarly, a player could have "earned" 36 points, a +7 compared to an actual total of 29, yet would be a relatively poor investment if his price tag was 12.0m. The point, of course, is that context is key, and that's what we hope to establish in the narrative below the chart.

With eight weeks in the books, it's just about time to look at some of the early over and under achievers and take a look at their underlying data to see where the model foresees them heading in the future. To get the most out of the below analysis, I recommend opening the player dashboard so you can play along at home.

Dousing the Fire

Yaya Toure -19
Yaya is a tough player to get a handle on as he does seem to outperform his underlying stats with some consistency (well. in his couple of years in England at least). That said, his goals per shots on target rate (G/SoT) still sits at almost 60% which is almost double what we'd expect and we'd forecast it to continue to regress as it has over the past few weeks. On the positive side, the current iteration of the model probably underestimates his bonus potential somewhat, given that he can impress the bonus system with passing, tackling and shooting (as opposed to, say, someone like Ozil who offers very little in the defensive half of the pitch). Currently we award 1.8 bonus points per goal scored (based on observed averages from the past two seasons) so this difference isn't going to bridge the gap between his points and expected points (xP) anytime soon. His xP of 33 puts him as a borderline top-10 midfielder and the certainty with which he produces adds to his value, yet at 9.5m we need to see an elite ceiling and floor and the data just doesn't suggest Toure can be that man. It's hard to sell on a high performing player owned by 37% of managers but all signs point in that direction right now.
Status: Sell unless you feel you strongly value consistency over value

Aaron Ramsey -19
Ramsey is enjoying an outstanding season and this ranking shouldn't be considered any kind of slight against his efforts to date. True, he's highly unlikely to continue to convert SoT at a 60% clip and created chances (CC) at a 43% rate, yet his production can take a huge hit and he'd still represent outstanding value. His xP of 39 is still 4th among all midfielders (Michu, Gerrard and Paulinho) and only Brady (whose value is largely wrapped up in penalties) and Mirallas can come close in his price range. Despite a trip to Old Trafford, the upcoming fixtures look reasonable and so current Ramsey owners should be happy riding the Welshman even if we expect his production to slow in the current weeks.
Status: Buy. He'll regress but has produced enough to more than justify his cost

Sergio Aguero -14
Aguero is quite possibly the best forward in the league and if there is one player whose totals I might be willing to accept won't regress all the way to a mean, Aguero would be in the conversation. Though high, a 50% G/SoT isn't insane for a player of his caliber, receiving the kind of quality chances he has, though of course some regression would be expected. We can also expect some decline in that shot on target rate (SoT%) which currently sits at 66%, up from 41% in prior year (for comparison, last season van Persie sat at 45%, Benteke at 38%). This could be partly offset by an overall improvement from City (who are averaging a full two shots less per game than prior year), but barring a dramatic shift, we're likely to see Aguero's incredible strike rate slow a little. With an xP of 42, Aguero has been excellent however you look at it, but at 11.0m+, some caution must be exercised by those looking to bring him in at the expense of the cheaper group of Sturridge, Giroud and Soldado.
Status: Hold. Has about as much upside as anyone yet hasn't produced enough to imply he's definitely worth the extra cost over his peers.

Loic Remy -14
Many will be surprised to see Remy on this list, not because we're saying he's overachieved but that to draw such a conclusion means the player has achieved something. Remy has received very little attention and just an 8% ownership number, despite notching five goals in his five starts this year for Newcastle and coming in at a reasonable price tag. He has, however, a couple of black marks against his name. First (as with most players in this list), his G/SoT of 63% is unsustainably high and almost double his historic rate of 32%. Second, he has been deployed in a deeper role in occasion with Cisse leading the line, yet has offered very little by way of assist potential, making him a one category fantasy player not playing in a position where he can best contribute to that category. We've seen players succeed in deeper roles before, so the latter point isn't a death sentence, but it is an issue when viewed together with his useful but not terribly inspiring xP of 28.
Status: Hold for this week against Sunderland but long term he could be considered for sale if better options emerge.

Adam Lallana -14
Another player who has received very little attention, Lallana comes out of GW8 with 44 points, a 4th placed midfielder ranking yet an ownership number of just 5%. Back to back double digit gameweeks will likely get the attention of the masses though, so the question as to the sustainability of his recent success becomes key. Though the model sees him as more of a 30 point player than one pushing the top of the leaderboard, several factors still go in his favour. First, his xP of 30 would still put him 11th among midfielders, with only two of those ahead of him (Brady and Townsend) available for a lower price. Second, no single factor in his underlying profile suggests steep regression is in his future. A 40% G/SoT looks a touch high but nothing to be worried about, though his A/CC% rate of 20% does look a bit high compared to his team's conversion totals to date.
Status: Buy. He isn't an elite midfielder but he doesn't need to be to produce value at 6.2m. Southampton have played better of late and have some promising fixtures on the horizon (starting this week with Fulham at home) and Lallana is well placed to take advantage of most good things his team does.

Oscar -12
Oscar is in a very similar position to Yaya Toure in that he's been excellent this year (xP 33) and could finish the season as a top-10 midfielder, yet at 9.0m his production is unlikely to offer excellent value. He hasn't proven to be a particularly clinical player in the past (largely driven by his percentage of long range efforts) so a 57% G/SoT is obviously not something to rely on. The bigger concern with him though is the issue of rotation as Mourinho has shown a tendency to shuffle his pack with some regularity. Oscar has been voluntarily benched twice already and with Champions League action in full swing coupled with Chelsea's excellent depth in midfield, we can expect that trend to continue over the coming weeks.
Status: Sell. Oscar is far from a bad option and his recent resting could suggest he plays the next couple of games. But, given that they include the visit of City and a tough trip to Newcastle, we could see Oscar essentially contribute nothing for the next month or longer if the regression monster really starts to kick in.

Daniel Sturridge -12
The question has arisen over the last few weeks whether Sturridge or Suarez is the best option in this promising Liverpool side and while the first reaction was Suarez, it is a very close call. This -12 score suggests Sturridge has overachieved, yet that's only because he's coming from such a ridiculously lofty position. Seven goals in eight games is almost unprecedented and only 39 players managed to score in seven different games all season last year, with players like Aguero (11), Hazard (9) and Rooney (8) not eclipsing that total by a wide margin. Sturridge's xP of 47 leads the league and the only real red flag in his data is his A/CC rate of 40%, though given that (a) not too much of his value is tied up in assists and (b) he has the potential to earn easy assists by squaring to Suarez, even that figure shouldn't terrify you. It should be noted that Suarez has posted outstanding numbers in his three game sample (earning an xP of 20 or 6.66 per game compared to Sturridge's 5.9 per game rate) but essentially both have the potential to lead the league in fantasy points and the key decision right now is whether you value a lower price (Sturridge) or a "differential" option (I use that term lightly given Suarez's 19% ownership number, but it's certainly preferable to Sturridge's 41%).
Status: Buy. Gun to my head, I still believe Suarez is the slightly better play, but given the lower price tag and statistical performance to date, you can't go wrong with either right now.

Steve Sidwell -11
Sidwell was highlighted in the preseason as a solid minimum priced option, though obviously it wasn't forecast for him to sit just outside the top-ten among all midfielders at any point of the season. In reality, of course, that lofty ranking is sure to collapse in the coming weeks with Sidwell representing one of the more extreme regression candidates in the league. The Fulham man has notched just four total shots with two on target (100% G/SoT), putting him in line with such goal scoring titans as Luke Shaw or James Tomkins. Incredibly, his A/CC rate is also 100%, a rate that will also plummet as the season progresses, no matter how clinical Dimitar Berbatov might be. 5.0m options are essentially a lottery and few, if any, will provide any kind of sustained production, so in that sense Sidwell's reliable playing time keeps him ownable. That said, for just 0.2m more, Brady is the current holder of penalty duties for Hull, a fact that should ensure he continues to pull away from Sidwell as the season progresses.
Status: Hold. Selling Sidwell isn't a priority but you need to be realistic about you have here.

Mesut Ozil -10
Ozil's dashboard is somewhat idiosyncratic, with a number of areas looking set for regression is opposite directions. He's actually performed below expectations in terms of G/SoT with a rate of just 29% compared to his historic average of 38%. On the flip side though, he's hit the target at an incredible (an unsustainable) rate of 88% which will cancel out most of any gains seen on his G/SoT rate. His A/CC rate of 33% looks high, even compared with Arsenal's excellent 19% average (the league sits at just over 10%), though the quality of his play and his teammates finishing ability could well see him finish the year at a rate well above league average. His 5.8 xP per game leads all midfielder to date and there's no metric to find that really suggests anything but a stellar season for the German playmaker, even if it doesn't finish with him notching a 7.8 PPG score.
Status: Buy

Fanning the Flames

Rickie Lambert +18
Lambert's numbers are fairly incredible to date, notching an xP of 44 (4th highest in the league) yet coming in just 15th among forwards in actual points. A low G/SoT should regress in the coming weeks and with a good share of Southampton's shots in the box, he should benefit from Southampton's improving play. The next three games look very promising and there are really no other reliable options in Lambert's price range who can offer the same level of upside and consistency.
Status: Strong buy

Wilfried Bony +16
Bony has been solid when he has featured for Swansea but has been slowed by missing a couple of games along with the unusual position of scoring two goals and an assist yet earning zero bonus points. Bony has accounted for 38% of Swansea's SiB when he's played which is an excellent rate and we'd expect some improvement from his 25% G/SoT rate, even if his historic rate sits at just 32%.
Status: Buy. At 7.5m there is value to be found here, especially given his penalty duties and upcoming fixtures, though between the two options above, the data gives a slight edge to Lambert.

Jon Walters +14
Walters has been better than his production suggests, hitting the target four times already without any returns. Yet, even if converting as expected, his xP of 28 would be just about in line with his price tag and not really offer a great deal of upside compared to his comparatively priced peers. Ownable? Sure. A key target? Not really, especially considering his team's attacking struggles.
Status: Sell, or hold if your team has serious issues to address.

Danny Graham +13
Graham has put in some decent displays for Hull without any reward, both in terms of goals and assists. His upside is limited, but at 5.1m he can be a useful bench option for those looking to minimise their expense on the bench. Unfortunately, Graham has picked up a knee injury and will likely miss a couple of games, but he can be considered for review once he returns.
Status: Monitor on his return from injury for increased signs of life.

Robin van Persie +12
van Persie's underlying numbers remain very good (xP 44) though a combination of a bit of bad luck coupled with that extortionate price tag have led to him being labelled a bust so far this season. The last three weeks have seen over half a million managers sell the Dutchman, somewhat strange timing given the next four promising games. Before the season I maintained that it was all but impossible for him to justify the extra investment needed to bring him in and despite his +12 mark here, the fact still remains that other players can offer similar returns for a much lower price. If timed right, I can see van Persie being a player to target when playing your wildcard; to try and milk as much value as possible over a short period, but unless we see a significant turn around from United, it's unlikely that he'll be the elite captain option every week, as he was last season.
Status: Buy in terms of production but the price tag would still scare me off.

Michu +12
Another of last year's elite is having some comparative struggles so far this year, though here we see a player who is really suffering through some unfortunate conversion rates. His G/SoT rates have been excellent throughout his career (in Spain and England) yet currently sits at just 20%, less than half of his 46% average. It's a similar story on the assist front, with his A/CC of 7% again coming in at around half of his team's average (13%). His xP of 46 simply leads all midfielders and is second only to Sturridge (47) overall and the upcoming fixtures look outstanding right up to Christmas, other than a tough trip to the Etihad in GW13.
Status: Buy. Not a true differentiator with an ownership number of 16%, yet he offers elite upside and a nice alternative to the overly familiar pair of Yaya Toure and Ozil.

Andre Schurrle +12
Given his price tag and classification as a midfielder, it would be great to add Schurrle to our shortlist, yet a lack of playing time and rotation risk makes him all but impossible to own for the foreseeable future.
Status: Monitor for playing time improvement.

Friday, October 18, 2013

Gameweek forecasts: a couple of case studies

Producing forecasts is a tricky business. Even with hindsight it is tough to predict the expected outcome of a given game (i.e. how shots transform into goals) and that problem increases exponentially when you also need to try and forecast the underlying data. Throw in uncertainty around how much players will play and the issue of small sample sizes and you have a recipe for some funky results over these early weeks of the season. First, to make sure the model isn't totally off track, let's look at how it performs retrospectively over the first seven weeks of the season (using actual shot totals as inputs):

Though we can see outliers in the above chart (especially at the top end of the market), the overall trend is promising and the r-squared of 57% for players with a risk factor of 2.5 or less is encouraging enough. That's not to say it's infallible, but it's a good start and for the majority of the extreme outliers we can point to specific factors which have led to their strong results (Yaya Toure is unlikely, for example, to convert his shots on target into goals at a 66% clip for the rest of the season).

So if we're happy that the model is working relatively well over a medium length period, let's get into the forecast side (which include predicting how many shots a player will get and how they will be converted) and look at a couple of actual examples from this week's forecast:

Loic Remy 7.3 points (you'll need to slide the risk slider to 2.7 or more to see him included)
Remy represents one of the dangers of forecasting in the early weeks; a problem that is compounded by the fact that Remy has only started four times. When he has played he's been nothing short of spectacular, averaging 4.4 shots per 90 minutes, hitting the target 50% of the time (a very useful rate). Despite this success, I imagine most people who asked about this forecast were confused given Newcastle's opponents. With just four goals conceded on the season, Liverpool have appeared to be a very useful defensive side and thus don't immediately jump out as a team you want to start your forwards against. However, digging a bit deeper we see a team with a +/- SiB rate of 22% away from home, surrendering the same number of SiB as teams like Villa and Cardiff (this despite playing AVL, SWA and SUN). This isn't to say, of course, that Newcastle are set to put Liverpool to the sword this week, but looking purely at the data, the Magpies' prospects this week are better than most would probably think (the model estimates them to notch around 9 SiB this week). With Remy accounting for 40%+ of his team's SiB, the model likes his chances this week, even if the intangible factors (playing time risk, wide role) suggest more caution.

Daniel Sturridge 4.6 vs Luis Suarez 4.1
If Remy is giving us trouble with five appearances, things get even tougher with Suarez who has played just twice. I suppose we could lean on prior year data but without wanting to reinvent the whole model for every idiosyncrasy, we'll just live with it. Long term I would personally back Suarez to top his English colleague in the scoring charts, though as this is a stat based site, I've nothing to base that on other than past events so it doesn't make it into the model. As always, these forecasts should support our decisions and if something seems off then we should simply ignore it.

Stevan Jovetic 7.3 (risk rating 3.9)
This is something of a "damned if you do, damned it you don't" situation as if I exclude players from the weekly listings I inevitably get questions where Player X is, and then when they're in, it seems ridiculous to rank the scarcely-used Jovetic as the top option. For clarity, all players make the listings and you can filter down to more reliable options using the risk slider. Jovetic's sample size is almost certainly too small to be particularly reliable and thus the 7.3 points ranking isn't wholly useful, but if you think this then simply ignore him. I prefer this approach rather than me deciding who is/isn't relevant and having readers miss out on a sleeper prospect they are targeting. This is the same approach used by publications like Baseball Prospectus who generate PECOTA forecasts for minor league players on the assumption they get major league playing time (and it's to the manager, player and luck whether the youngster gets his shot). For what it's worth, in his limited time Jovetic has been excellent (6 shots, 4 SiB and 2 SoT in just 85 minutes) and thus the model likes him to succeed were he to be given time in this talented City side.

Julian Speroni 5.7
This is probably the strangest forecast for the week and I must admit, I went back to recheck the data myself before posting these numbers. The fact that it's Speroni (and hence Crystal Palace) is somewhat surprising though when you consider their opponents' pathetic efforts away from home to date, it doesn't seem overly unusual to see Palace ranked well this week (indeed they would rank 3rd in the standard "goals per game" projections from last season). The issue then, is more that a goalkeeper is the week's highest ranked player with a low risk rating which doesn't feel quite right. Part of this is simply the perception of fantasy players and part of it is a lack of sophistication with the model.

On the first point, Artur Boruc ranks 9th among all players this year and three 'keepers place in the top-15 so in reality it isn't particularly surprising to suggest a 'keeper will score well (remember that they tend to earn more points than defenders per clean sheet due to the presence of saves). They lack the upside of defenders who can also score or notch assists, but for most players the chance of those events is fairly low and thus doesn't have a huge factor on a weekly ranking. The second point is a bigger issue and that's how 'keeper points are calculated. Right now, saves are awarded on a very crude average basis which doesn't take into account the propensity for earning saves in a given game. Thus, while Speroni has racked up decent save totals to date, if his chances of a clean sheet are higher this week then his save totals should go down, but the model doesn't make such an adjustment. This is unlikely to result in more than a half point variance in a given week though.

The wider point on 'keepers and to a lesser degree defenders is that the forecast is more likely to be wrong on them in a given week as their chances of success is a little bit "all or nothing" (where as midfielders can earn points for clean sheets, assists, goals and are more likely to get the bonus nod). I still feel fairly confident is using this data for ranking purposes but wouldn't suggest captaining Speroni (5.7) over someone like Michu (4.9) given the exponentially higher upside enjoyed by the Swansea man. To repeat, this data is based on logic but a simple model cannot account for every possibility so personal judgement is still required.

Hopefully these examples gave a bit more detail to how the weekly rankings are made and we'll continue to check back to see how they are performing as the season goes on (we'll hopefully see some of the stranger outliers disappear as sample sizes start to increase). Thanks for reading and for sticking with the blog during the quiet opening weeks and please continue to send your questions to @plfantasy, on Facebook or in the comments below.