Friday, January 18, 2013

Cheap Links

Sometimes I hesitate writing certain posts as the content of them sounds somewhat obvious. That said, doubling up on double gameweek players or selling those who have lost playing time all seem self evident but are continually ignored so I guess nothing is too obvious.

The chart below plots the starters from each team defence's price against their ownership, hoping to identify the 'cheap links' within a given side. The logic being, of course, that the majority of most defenders' points come from clean sheets, which are earned as a team, rather than as an individual. As with any decision, it isn't as simple as finding the cheapest defender from a team and throwing him into your side, as we also need to consider their job security, and of course their attacking potential. I'm not going to dwell on the later, suffice to say that in many cases we are often guilty of overpaying for that attacking threat, which is generally extremely hard to predict. There are exceptions (or perhaps, an exception with Leighton Baines) but my general view is to air on the side of caution when forecasting attacking points for defenders.

The size of the square on the below chart represents a players' attacking potential based on their underlying stats to date. Given that we're close to the halfway point of the year (and thus everyone has essentially played everyone once) I have not adjusted the forecast for opponents, so it should be taken with a pinch of salt, merely as a quick and dirty indicator of wide gaps between two given options.

I have briefly touched on every team, but for many there really isn't much to say. This piece is specifically about cheap links in a given team so I will not dwell on debating the "merits" of the Villa defenders, all of whom essentially cost the same and offer little by way of differentiation (or in truth, points).

In the below projected lineups, players who I deem to be safer are marked in bold. In many cases I have little to base this on other than historic playing time, so if you feel someone is misrepresented, my apologies, and please adjust them in your own rankings.

Arsenal
Projected lineup: Gibbs, Mertsesacker, Vermaelen, Sagna
This one might be one of the easiest picks in this piece and to me at least represents free money for those currently holding Gibbs, Vermaelen or the suddenly less secure Mertesacker (he was benched in favour of Koscielny this week, who was of course sent off, but long term you have to now question the German's place in this side a bit). The attacking drop off from Vermaelen to Sagna is non existent and the narrative of Vermaelen as some kind of offensive centre back has simply not played out in the stats. He's more likely to get an entry in a goal of the month competition but that privilege doesn't come with a fantasy bounty. As for Gibbs, he has put up better attacking numbers than Sagna, but probably not to the point of justifying the extra 0.5m investment. Sagna has historically (and this season) shown an ability to chip in with one or two goals and 3-5 assists and there isn't much here to suggest Gibbs will significantly eclipse those totals this year. Throw in the fact that Sagna offers a nice differential option from your rivals and he makes one of the better defensive buys around.
Pick: Sagna, definitely over Vermaelen and Mertsacker, and just ahead of Gibbs

Aston Villa
Projected lineup: Lowton, Vlaar, Clark, Baker
As noted above there's little value in this per se, and with no one standing out as sufficiently cheap to get involved with, let's move on.

Chelsea
Projected lineup: Cole, Terry, Ivanovic, Azpilicueta, Luiz (midfield)
If it wasn't for the question mark over his playing time, Azpilicueta would be by far the best pickup here and I'd be yelling his name from the roof tops. He hasn't brought a great attacking threat but he does enough and it's unlikely that over the next 16 weeks Luiz does enough offensively to justify the extra 1.0m it'll take to get him. That said, he faces competition not just from Luiz (who could slide back into defence from midfield at any time) but also from Terry and Cahill who are now both fit and ready to go. If I had to guess I'd suggest that Ivanovic is preferred in the middle and Azpilicueta - for me a more 'typical' full back - is kept at right back. If you're team is really limited for transfers Azpilicueta might represent a risk, but if you're wildcarding or have the depth to cover him should he dropped, he represents a very solid risk-reward candidate.
Pick: Azpilicueta as a somewhat risky upside play

Everton
Projected lineup: Baines, Distin, Jagielka, Coleman
This is a tough one as while my instinct is to shun the pricey Baines for the reasonably priced Distin, his production and potential is undeniable. In the admittedly crude expected points calculations used in the chart, Baines forecasts to score 40 attacking points for the remainder of the year with only Assou Ekotto (31), Evra (25) and G Johnson (21) managing to top 20. Given that the Spurs man's number is based on a tiny sample size we basically have a situation where Baines is almost twice as good as all other defensive players. Is that enough to justify the extra 2.2m over Distin to enough to compensate for Everton's lack of defensive prowess versus the likes of Man Utd (Evra) or Chelsea (Luiz), I'm honestly not sure. That same 2.2m could be the difference between Nolan and Bale, or Podolski and Suarez, gaps which have been similar as the Baines-Distin spread to date, but which look (a) more likely to continue given Distin's increased playing time and (b) are subject to more upside given the form Bale and Suarez have shown. As for the gap to other more projectable defensive sides, the model is really not a fan of Everton's upcoming schedule and thus it's hard to say that Baines is a sure bet to eclipse the likes of Evra or Luiz, hence my preference for grabbing Distin in this side
Pick: Distin looks like the safest pick though Baines' attacking upside is legit and it's at least arguable that he can justify that premium if your squad value is relatively high

Fulham
Projected lineup: Riise, Hughes, Hangeland, Riether
Fulham are a solid yet unspectacular defence who should be counted on more as good rotation defenders rather than ones you can deploy every week without question. That means price becomes a big factor (provided you have guaranteed playing time which probably applies to everyone here despite Riise's recent stint on the sidelines) and thus Hughes is the only player I'd touch here with Baird no longer a stable option in midfield. Hangeland and Riise just don't bring enough to the table to warrant the extra cost and given their 5% and 6% higher ownership, this moves looks like a no brainer to me.
Pick: Hughes

Liverpool
Projected lineup: G Johnson, Agger, Skrtel, Wisdom
This decision may be the easiest of all, with Jose Enrique sidelined for another month, thus giving Wisdom another four or five games in the first team, including home against Norwich, West Brom, and hopefully Swansea. If you're not wildcarding, or have already done so, I guess there is a question as to whether you'd want to burn two transfers on a player who also faces two games in which you'll have to drop him (@ARS, @MCI) but otherwise he looks like one of the best pickups available.
Pick: Wisdom, by a long way (please also note the lack of use of the hacky "consensus Wisdom" pun here.

Man City
Projected lineup: Clichy, Kompany, Nastasic, Zabaleta
There's really no value in this back line in terms of cheaper assets, so you really just have to try and balance out their job security and price tag, which right now would probably lead you to Zabaleta. That's another mess for another day though.

Man Utd
Projected lineup: Evra, Vidic, Ferdinand, Rafael
It's the same story on the red side of Manchester as you're really just hoping for your United players to play, with no one coming at an extreme discount. Evra and Rafael look closer in terms of attacking potential than their points haul would suggest, though Rafael does also face a slightly higher chance of rotation with P Jones waiting in the wings. Again, this mess is not really the focus here, so let's keep going.

Newcastle
Projected lineup: Santon, S Taylor, Coloccini, Debuchy
Once S Taylor returns he looks like he'll be the best play here, coming at the lowest price, owned by no one and bringing a much better attacking threat than his tiny sample size for this season suggests. Moving Santon to him doesn't look like a priority but Coloccini owners should probably take a longer look given the added presence of his differential ability.
Pick: S Taylor, when healthy, looks like a worthy play for a slight discount over his teammates

Norwich
Projected lineup: Garrido, Turner, Bassong, R Martin
Bassong and Martin have each enjoyed brief runs of big points production, but neither have the underlying stats to suggest this kind of form is here to stay. Saying that, Whittaker out indefinitely (Martin's main competition), his attacking upside looks to be be sufficiently high to warrant the higher ownership percentage so I'd take him over Turner right now (though wouldn't be in a rush to make the move either way). Bassong owners should strongly consider cashing in and saving themselves over a million pounds given his much higher ownership numbers and limited attacking upside.
Pick: Martin

QPR
Projected lineup: Traore, Hill, Mbia, Nelsen
There's really no point getting involved here until we know what Redknapp will do in the transfer window. That, and the fact that this team is somewhat devoid of fantasy talent.

Reading
Projected lineup: Harte, Mariappa, Pearce, Gunter
Everyone can be had for under 4.0m so you're free to pick whichever one of these cheap options will disappoint you after showing flashes of promise during their three clean run in GW18-20.

Southampton
Projected lineup: Shaw, Hooiveld, Yoshida, Clyne
Clyne and Shaw look like the best plays here, with Clyne probably getting the slight edge despite his higher ownership numbers thanks to a better attacking threat. We're less bothered about that higher ownership percentage for players like Clyne, as the majority of those owners won't be playing him so any upside would be felt by few of those ~6%.
Pick: Clyne

Stoke
Projected lineup: Wilkinson/Wilson, Huth, Shawcross, Cameron
Marc Wilson is back is training and will represent sensational value at 4.0m when he's fully fit. Until then Wilkinson remains the best pick here as Huth and particularly Shawcross's attacking threat is really overrated and you're really only paying for Stoke's very strong defence. With Wilson possibly back next month, if you were using a transfer I might lean towards for Cameron for safety, but if wildcarding I'd feel comfortable slipping Wilkinson in here and then making the move to Wilson when he regains his full fitness.
Pick: Cameron in the short term but Wilson for the long term when healthy

Sunderland
Projected lineup: Colback, O'Shea, Bramble, Gardner
With Cuellar sidelined, Bramble becomes the cheapest option here but that move comes with a downgrade to this unpredictable defence so I wouldn't want to get carried away. However, 4.3m for a player who looks pretty much guaranteed playing time until Cuellar's return is solid value so Bramble deserves a look as a 5th defender type.
Pick: Bramble

Swansea
Projected lineup: Davies, Chico, Williams, Rangel
Rangel is presenting with a much stronger attacking threat than Davies, and it's to a point where the 0.5m investment may almost be worth it, especially considering the higher risk that the young Davies will face more time on the sidelines (replaced by the very competent Tierdalli) than the more experienced Rangel. If you have money lying around it's probably worth the upgrade but I'd be more than okay holding both of these options and would swap Williams for both.
Pick: Davies, or Rangel if you have spare cash

Tottenham
Projected lineup: Assou Ekotto, Vertonghen, Dawson, Walker
The three pricey options here all look like reasonable options though are pretty much priced where they should be. Dawson on the other hand could be one the best value defenders in the whole game, if you believe he can hold off the threat from Caulker and Gallas. I probably wouldn't suggest he'll play every game but I believe he'll play enough to be an excellent rotation option and would strongly target him in the next few weeks.
Pick: Dawson

West Brom
Projected lineup: Ridgewell, Olsson, McAuley, Jones
Jones' price tag is promising and the playing time has been pretty solid of late, suggesting he's the pick here. He does come with a slightly lower attacking threat than Olsson which together with slightly less job security might lead one to conclude that Olsson remains the pick here. I'd venture it's pretty close and neither look like options you'd really want to target right now.

West Ham
Projected lineup: Demel, Collins, Reid, O'Brien
With McCartney sidelined and O'Brien's price rise, Demel once again becomes the cheap link here and should represent solid value at 4.0m. O'Brien owners who got in at a lower price will likely feel okay staying put but Reid owners should think about this move as the 1.1m difference in price can be better used elsewhere.
Pick: Demel

Wigan
Projected lineup: Boyce, Ramis, Figueroa, Caldwell, Beausejour
It isn't particularly clear who will play in the Wigan side, nor in what formation they will start and thus with no super cheap options there's really nothing to focus on within this whole group.

6 comments:

FC Buckingham said...

I like the fresh approach, but I think I am missing something.

If team clean sheets, predictable playing time and price are driving the decision, wouldn't it make more sense from a differential perspective to look at % of FPL players who have the defense rather than a particular player? If you have Sagna at 2% and everyone in your mini league has Gibbs at 10%, when Arsenal gets a clean sheet you haven't really differentiated yourself at all.

Considering your valid view on the unpredictability of attacking points, I think an interesting approach would be to look at "Team Defense Ownership %" (sum of likely starter's ownership %), team clean sheet probability over the next 6-weeks, then identify the cheapest locked in starter on each squad to try and identify differential value. With a strategy that you transfer a new differential defender in when the value shifts due to fixtures? Or is it just not worth it to use transfers on 4m-5m defenders?

CDI said...

Great stuff as always Chris.
I'm interested in your opinion on Ba and how much rotation he'll face going forward. Do you think its enough to warrant the Ba-> Podolski move thats so popular this week? I got Ba at 7.7 so there will probably be no going back if I decide to get rid of him now.

Other options for the week are Azpilicueta -> Sagna. I already have Gibbs but its a DGW and arsenals fixtures are not that bad going forward. Which of the 2 options would you pick if you were in my shoes? (Feel free to say neither if they both suck). Thanks.

CDI said...

I forgot you did a piece on Ba but has your opinion on him changed since then? He started both games in the DGW and when Torres came on he looked abject. My mian worry is that Rafa won't simply send Torres to the Horse farm and will be forced to give him mins. Whether thats with Ba or at Ba's expense remains to be seen.

vanilla said...

@FC Buckingham:
You have a point about the "team ownership" piece, but as to the differentiation between Gibbs/Sagna, I think the point of this piece answers your question. Chris is looking for the best use of budget to get into any one of these defenses. Which defense to pick is up to you. Picking Sagna over Gibbs doesn't earn you any points from that position, but the 0.5m you can reinvest elsewhere may have.

vanilla said...

* "team ownership" comment (not piece)

Chris Glover said...

FC Buckingham - yeh good point, I should have made that clearer. The differentiation I am suggesting is that if Sagna and Gibbs offer the same defensive potential and their offensive upside is essentially a wash, I'd prefer to (a) take the cheaper guy and (b) take the guy owned by less managers so his points make a bigger impact.

I do like the idea of looking at which teams are owned, the tricky point being of course that someone like Reading might have an inflated number because they offer a number of ~4.0m options who rarely see the light of day off players' benches.

The wildcard piece I did laid out my plans and I am indeed doing Ba> Pod this week. Ba may well work out but for the price and the availability of options like Pod (or longer term guys in a higher bracket like Aguero or Rooney), I'm not willing to risk him sitting. While some players *might* get dropped, Ba simply has to, as I just don't see a scenario where Torres doesn't play 40% of all games. I guess Ba could slide out wide now and then, but it's not like CHE are starved of talent their either. I fear CHE options may become like MCI players whose talents are tempered by the constant threat of rotation.

I like both Sagna and Azpilicueta long term, but on a one week scale, Sagna would be a clear upgrade for the week per the model. Of the two transfers I might lean towards getting Podolski, but only if you then have the next step planned as I wouldn't forecast him to be great value over a longer period (good, but not great)

Vanilla - exactly right.