Topic: RAWK Round Table Hull City AFC 3-1 LFC

[-Daws-]

Off memory, Suarez much more so than Moses.

Didn't look anywhere near fit. I'd say the absolute battering he got at the bitters would have something to do with that.

[PhaseOfPlay]

Off memory, Suarez much more so than Moses.

Suarez - one successful defensive action in the opposition half - Moses, none. Moses - 6 defensive actions in the defensive half, Suarez - none.

Henderson was our best pressing attacker, by far. He went beyond both Moses and Suarez at times.

[PhaseOfPlay]

Forgive my ignorance, but what is included under "defensive actions"?

Also, the absolute number of actions itself can be misleading, no? If the other team has 0, 1, or max 2 passes in their half before 'launching it' then the opportunities for high pressing are reduced.

Anyway, I may be grasping at straws.

Tackles, interceptions, second ball recoveries.

And if the opportunities or high pressing are reduced, then we're not really high-pressing, are we?

[MobileBayRed]

We only had 11 defensive actions in Hull's half

That was 11 more than we had in our own half. . . .

I'll get my coat.

[-Daws-]

Suarez - one successful defensive action in the opposition half - Moses, none. Moses - 6 defensive actions in the defensive half, Suarez - none.

Henderson was our best pressing attacker, by far. He went beyond both Moses and Suarez at times.

At times? It seemed every time their centre halves or goalkeeper had the ball Henderson was the further man forward chasing the ball relentlessly. Suarez looked dead on his feet and as usual Gerrard and Lucas struggled on their own away from home.

[007.lankyguy]

That was 11 more than we had in our own half. . . .

I'll get my coat.

Nah, I had a little smile at that mate

[John C]

IMO, that´s totally wrong mate, it´s similiar to the hype in regards of Toure IMO. Flanagan isn´t anywhere ready to be a regular, Cissokho is still a level above him in terms of everything you can think of. I think Flanagan was given game time by Rodgers because he wanted to reward his effort in training, similar with Toure but that´s not working on a PL level.

You mustn't misunderstand me mate, I'm not over-hyping Flanagan, I find the theory of doing that to an unproven player as infuriating as you may do. I'm just suggesting he has as much right to stake a claim based on performances as Cissokho.

I'm holding back on my opinion of Cissokho and I always rated Robinson more than Flanagan

[PhaseOfPlay]

That was 11 more than we had in our own half. . . .

I'll get my coat.

LOL

Cruel but true

[robertobaggio37]

Bad news for the alcoholics among us then.

[Trendisnotdestiny]

I would seriously consider going back to a 3-5-2. From a management standpoint, either Rodgers has to tell one of the defenders they are being moved on, or play the back three with one spare. Having two spare defenders seems to be causing him indecision, and this indecision is filtering through to the team. He also has to sort out midfield, and that means confronting the Gerrard problem. Ultimately, this will be the biggest hurdle to get over. He is the club captain, the club legend, and his set-piece delivery has been back on form this season. But in open play, he is probably the most guilty player of all in terms of changing our style of play. Not because he plays long passes - he plays less per game so far than last season - but because he isn't responding to the needs of the game. When we're looking a bit sloppy in the attacking third, then you abandon the through ball and work it from side to side try to create an overload. That creates the space to drive through the middle on the dribble. Either way, you need your main midfield distributor to put their foot on the ball and allow players to get into position. Instead, today, without Sturridge, we were still playing the long ball even though we didn't have the players to make the most of them, and Suarez is not a target player. Aspas might have been a better option if he was further along in his recovery, than either of Moses or Sterling, but this game was about sloppiness and lack of effort, and the long balls didn't allow us to get a grip on the game.

Conversely, the defenders HAD to play the long ball today, because they didn't have the personnel to take the ball into midfield and create something, and there was no movement ahead of them. Agger and Sakho, of course, were prominent in their absence. In the pre-match thread, I responded to some posters concerned about the physicality of Hull and their set-pieces, because I thought they were not something to worry about as much as people thought. Apparently Rodgers sided with that frame of mind, and it looks like he selected the central two with that as a concern. But that was a very cautious way of thinking. We needed to be aggressive in our intent, and force Hull to deal with us. Instead, we tried to nullify the game completely by starving them of the ball, but that only works when all 11 players are switched on to a possession game, and are willing to work for it. Our players weren't.

In the end, we were left looking starkly at a number of problems that have been mentioned throughout the season, and which Rodgers needs to address very quickly, before they become an albatross around the mananger's neck. Those problems are:

- Gerrard doesn't have the power to play box-to-box any more, although he insists on doing so. In doing so, he becomes unable to track his man

- Sturridge is out for 2 months, worse case scenario

- The fullbacks are off-form, and Enrique is injured

- We have a surfeit of central defenders, and only two of them can bring the ball into midfield

- We are vulnerable through the middle, still, to direct attacks when we play a back four

- We need to get Coutinho on the field as much as possible

For me, this all points to one viable solution, that at least alleviates all of the above problems: we have to return to the 3-5-2. Sure, we conceded goals with it, but we also created more and scored more - and as of now, we've lost less goals than with a back four. It solves the defender conundrum because we can play out two best central defenders plus one other, in all games. It solves the forward conundrum because we can still play two forwards, but we can also experiment, as someone mentioned, with a Gerrard and Suarez combination where both players are shorn of the majority of their defensive duties. It gets Coutinho on the field in a midfield three. And it gives the fullbacks the support they need to be able to push forward aggressively the whole game. Sure, we'll end up as a 5-3-2 in some games, but we looked a lot better in this formation than we have in any of the back four versions this season, for my money.

This game was not really lost on tactics per se. We lost it in the selection of the starting team, the formation they played in, the lethargy of the players, but also the massive gap left by two of our most limited players (one technically, the other by age) at the back and the lack of thrust up front. A temporary 3-5-2 with Gerrard up front would give both of those players free reign to worry about goals, give Coutinho mobile targets to play off, and give us the chance to see a more balanced central midfield in front of a defensive arrangement that gets the most out of the defenders we have, rather than Rodgers trying to pick 'n' mix the four starters according to the games, when really there is probably only one combination of two that works. For Norwich, we could go for something like this:

Mignolet
Skrtel----Agger----Sakho
Johnson---------------------------Cissokho
Lucas----Allen
-------Coutinho
Gerrard
---Suarez

Or

Mignolet
Skrtel----Agger----Sakho
Johnson-----------Lucas--------------Cissokho
Gerrard------Coutinho
Aspas---Suarez

Rodgers should be looking at maximising the team for the good of the league campaign, rather than trying to shoehorn players into a formation that is unbalanced. That won't prevent lethargic displays, of course. But it might put the players into a shape which plays to their natural strengths, gives them better reference points for attacking, and force other teams to adapt to our shape, rather than us adapting to theirs.

Phase of Play,

Good analysis.  Agree with 3-5-2.  However, I might push Glen over to the left and play Henderson at wing back on the first lineup you described.  Maybe even sit Gerrard and play Alberto next to Suarez. 

[Garlicbread]

When Moses got that really good chance that McGregor saved, was that at 1-1 or 2-1? Could have changed the game completely if it was at 1-1.

[Trendisnotdestiny]

Our defensive woes are centred on two areas - set plays, and central midfield. Unless Rodgers fixes those, he will find himself in trouble later on. 

I could not agree with this statement more.  And as you have said repeatedly, both of these problems are not difficult to fix.  Successful repetition breeds good re-start mentality.  And we need to account for the limitations in our center of the park (mobility of Gerrard & Lucas, size of Allen, and an overall lack of pace).   I do not see this as rocket science.  This leads me to the idea that:

If we shore these two up
Find one more scorer (from team or transfer market)
Get settled into a dependable line-up and formation

We should be fine. 
*** The one exception is if Glen's form continues to slide. 

[PhaseOfPlay]

When Moses got that really good chance that McGregor saved, was that at 1-1 or 2-1? Could have changed the game completely if it was at 1-1.

1-1

They scored a minute later

[Sangria]

I could not agree with this statement more.  And as you have said repeatedly, both of these problems are not difficult to fix.  Successful repetition breeds good re-start mentality.  And we need to account for the limitations in our center of the park (mobility of Gerrard & Lucas, size of Allen, and an overall lack of pace).   I do not see this as rocket science.  This leads me to the idea that:

If we shore these two up
Find one more scorer (from team or transfer market)
Get settled into a dependable line-up and formation

We should be fine. 
*** The one exception is if Glen's form continues to slide. 

We can do both these things in one go with the players we already have. We've just shown no indication that we're going to do so.

[Bobinhood]

I'm more inclined to think the "indecision" is simple rotation.

CB Rotation has been awkward for Rodgers because he's told everyone Skrtl had won his place, which he did, and he had to wait for him to make a mistake at least before he rotated him just on face value. so that took 5-6 games but now he threw up 2 ogs in one game so can do whatever he wants now. Probably Sakho will be in tomorrow. SAggerkho , probably. If so, that's a simple once through rotation at the earliest opportunity. Nothing to cry foul about.

Leads to an unsettled back four by definition, but they are all good players anyway. Id rather have a settled four myself but anyway I think its purposeful not indecisive. For me he LB situation is much more urgent Johnson on auto-pilot and no Enrique and suddenly its all so different then what we had in mind. If Cissokho ever wanted a chance to seize he couldn't ask for better.

btw I thought skrtl was immense v hull otherwise, shit game as it was anyway. Didn't read the whole thread not sure what the general take on that is, probably different I suppose.

[Robinred]

Here is the one figure you have to look at. Forget TSR, which is what I think you're referring to (although it's a good metric, I think even Grayson has some misgivings about it?).

I wonder what Mr. Cholmondely-Warner has to say about that?

http://www.youtube.com/watch?v=y4CXY6TVBMc

[PhaseOfPlay]

I wonder what Mr. Cholmondely-Warner has to say about that?

http://www.youtube.com/watch?v=y4CXY6TVBMc

"L is for Lice. L is for Labour"??

[GrkStav]

I just finished watching the 2nd half again. There was practically no forward pressing. And our play was a LOT slower than it was in the first half.

Most of the slowness of our tempo was not by choice, and it wasn't of the deliberate, get-them-slowly-disorganized, prodding and retaining possession type. I think it was due to Hull's tactics, and personnel deployment. After they scored the 2nd, we had more of a free reign in midfield and did manage to string together some 'moves' in a somewhat deliberate fashion.

Big, big missed chance by Moses, although it wasn't a gimme.

[GrkStav]

All in all, though, I find the relative slating of Gerrard and Lucas defensively to be exaggerated, especially of the latter.

[blert596]

I keep banging on about this but until we get a midfield that actually looks like it might want to become part of the attack then were going to need both Sturridge and Suarez on top of their games.

Defensively, apart from Lucas, our midfield is pretty much a mess. For all Hendersons pressing, unless an ageing Gerrard and whoever else is next to him plays that game then its unfair to expect Henderson alone to do it - and then whinge when he doesnt contribute enough going forward.

Attacking wise from the midfield we're pretty fucking hopeless. Either Suarez or Sturridge do something special, or our only other alternative is to hope Gerrard whips in a ball. Or PC manages to avoid getting kicked long enough to find one of a total of 2 (now 1) players who look like they could score.

Our midfields not good enough defensively or offensively.

We need to bring in Allen and let him have a few games. At least try the famed ball retention.

I also think the shape of the team needs to be decided on based our defence. Either sort out 3 at the back or go to 4 and stick with it. As long as it involves Agger and Sakho then its an improvement. Skrtel And Traore are excellent backups to have.

Lastly I think the manager needs to start laying down the law a bit in telling players they'll play HIS way, or they wont play at all.

[ElstonGunn]

Yeah our TSR is bad.... but then so are other measures Sotr, fenwick and PDO. Well not bad but mid table as opposed to where we want to be

The last 2 games have been bad however I've been following the numbers all season and it's been pretty consistent unfortunately - even when we've been winning. Basically our attack compensates for a lot.

However you do make some really good points and I don't want to over do the drama - we have things to improve rather than being in crisis
I'm not as worried about the central defence as midfield as I think in defence we have the personnel to improve but need better organisation
Both the numbers (the many shots conceded in total but doing slightly better in not conceding them from inside the box) and the eye test point to us having a huge problem off the ball in central mid and I'm not sure how we fix that with the current personnel

The "SAS" (bluergh) part of the season is over for the time being and we probably did make some defensive sacrifices to make that work - however we reorganise now we need to be better off the ball now as the attack won't be able to compensate in the same way

This is all true, though as far as I can tell, the more the model takes shot quality into account, the better we look. MCofA's, for instance, which is entirely based on an expected goals model that heavily relies on shot quality, has us as more or less even with the other top 4 hopefuls. The worrying thing is that while predictiveness of offensive shot quality is high, the same isn't really true of shot quality conceded. Anyway, leaving that last piece aside, I think we're actually looking a littler better that your account makes it seem, we're in the top 2 or 3 for shot quality, both offensively and defensively, and a lot of models haven't really incorporated that data yet. I'd say we're top 4 offensively, and mid table defensively. Definitely needs improvement, as you say.

[woof]

Had Suarez had one of his better days and got us out of jail, we wouldn't be this animated with the soap dish. However, it will still mask the deficiencies we have in midfield AND after a third of the season has gone, we still do not know which is the best CB partnership. These are big big concerns

[GrkStav]

I think we know that Toure-Skrtel is not our best CB partnership. And we knew that even before the match vs Hull City.

We also knew that a '2-1' midfield does not work best with Lucas-Gerrard being the '2'.

I reckon BR's plan was for Moses and Sterling to occupy their wingbacks. It didn't quite materialize, did it.

And yes, Suarez didn't have his best game.

Btw, for Hull City's 2nd goal, Sagbo clearly fouled Toure (pushed and/or obstructed). Of course, so did Skrtel originally, but be that as it may.

[Arcadian]

Why do professional athletes/sportsmen expect others to just roll over and capitulate?

We don't know how to fight. Or rather we don't care enough to fight week in, week out.

The side we fielded should have beaten Hull. True story. Take that team on their game and tell me that it's not good enough to beat Hull City...

Analyze the performance till you're bleeding from the retina. But we gave it up and THAT'S what has us raging.

[adityasharma14]

I would seriously consider going back to a 3-5-2. ....

Mignolet
Skrtel----Agger----Sakho
Johnson---------------------------Cissokho
Lucas----Allen
-------Coutinho
Gerrard
---Suarez

Or

Mignolet
Skrtel----Agger----Sakho
Johnson-----------Lucas--------------Cissokho
Gerrard------Coutinho
Aspas---Suarez

Here's my problem with a 3-4-1-2, especially the Rodgers sets it up.... it is basically structured to do one thing very well: win the ball quickly with a high press, and get the ball into the feet of the front two as quickly as possible. He basically used it because in Coutinho and Allen's absence, there is no one in midfield who is really good at linking midfield and attack. So you use a high-press, win the ball real quick and then try to use overloads to get the ball into the front two's feet real quick. There is a really really high dependency on the front two, is what I'm trying to say. With Sturridge out of the side, I can't see Aspas / Alberto / Gerrard filling that role any better than they would with 4 at the back.
Furthermore, the full-backs are extremely crucial in such a setup. Johnson would work on the right, but the left flank would be badly compromised if the only available options are Cissokho and Flano.

Lucas, Allen , Coutinho, Alberto: all our midfielders are better suited for a pass and move game more than a high-pressing game, as none of them have the kind of legs and stamina needed (Hendo does, but he's not good enough when pushing forward).

[The Woolster]

By the way, especially those of you who have access to the more sophisticated match-analysis software etc:

I am under the impression that we actually did attempt, on numerous occasions vs Hull City to press high. We just weren't that successful.

Is there a way to establish empirically, but in summary form, whether we did in fact attempt to press high on a regular basis?

Hopefully this link will work

http://www.fourfourtwo.com/statszone/8-2013/matches/695025/team-stats/14/OVERALL_04#tabs-wrapper-anchor

Have to say that whilst the number isn't particularly high, going from memory, the events were happening further up the pitch than usual.

[The Woolster]

Yeah our TSR is bad.... but then so are other measures Sotr, fenwick and PDO. Well not bad but mid table as opposed to where we want to be

The last 2 games have been bad however I've been following the numbers all season and it's been pretty consistent unfortunately - even when we've been winning. Basically our attack compensates for a lot.

However you do make some really good points and I don't want to over do the drama - we have things to improve rather than being in crisis
I'm not as worried about the central defence as midfield as I think in defence we have the personnel to improve but need better organisation
Both the numbers (the many shots conceded in total but doing slightly better in not conceding them from inside the box) and the eye test point to us having a huge problem off the ball in central mid and I'm not sure how we fix that with the current personnel

The "SAS" (bluergh) part of the season is over for the time being and we probably did make some defensive sacrifices to make that work - however we reorganise now we need to be better off the ball now as the attack won't be able to compensate in the same way

TSR, SOTR, and Fenwich are all limited and have the same issues, they take no account of chance quality, and as we are giving away a lot of shots from distance we will look relatively bad by those metrics. We are actually one of the better teams when it comes to PDO though, The reason for that is that we are creating good quality chances, so have a high scoring rate, and are giving up difficult chances, so have a very high save ratio.

When you start to include chance quality into the metrics, then we are doing ok. We are probably between 4th and 6th by those, and we are doing better than that in the league because Mignolet has saved more than expected (although that might have changed over last 2 games) and we have finshed our chances better than expected as well. Our scoring and save rates might regress, but we our underlying numbers are probably on target for about 4th, which is our realistic aim.

The other issue with those metrics is that they do not take into account game states. Teams that are 1 goal up tend to have a lower TSR as they start to shell and the losing team ries to get back in the game. We are only behind Arsenal when it comes time spent winning this season, so ratios like TSR will naturally be a bit lower than perhaps expected.

[ElstonGunn]

TSR, SOTR, and Fenwich are all limited and have the same issues, they take no account of chance quality, and as we are giving away a lot of shots from distance we will look relatively bad by those metrics. We are actually one of the better teams when it comes to PDO though, The reason for that is that we are creating good quality chances, so have a high scoring rate, and are giving up difficult chances, so have a very high save ratio.

I totally agree with your post, but I think the comment on PDO is misleading. Our PDO being high is actually a bad thing from a predictive standpoint, since it usually regresses to average. Although more nuanced research has found that a relatively small variation in PDO can be down to skill, it's still only worth 6-8 goals a year between the best and worst. I still have a hard time accepting these findings--shot stopping, finishing quality and the likelihood that better shots on target are more likely to be goals only accounts for a max of 8 goals??--but that's what the numbers seem to say.

[The Woolster]

I totally agree with your post, but I think the comment on PDO is misleading. Our PDO being high is actually a bad thing from a predictive standpoint, since it usually regresses to average. Although more nuanced research has found that a relatively small variation in PDO can be down to skill, it's still only worth 6-8 goals a year between the best and worst. I still have a hard time accepting these findings--shot stopping, finishing quality and the likelihood that better shots on target are more likely to be goals only accounts for a max of 8 goals??--but that's what the numbers seem to say.

Personally, I'm not a massive fan of PDO being a measure of 'luck' simply because it regresses a lot. The same issue with chance quality that TSR et al I think can still apply, but no ones really looked at it. The first attempt that I have seen was only in the last week or so here http://thebubblegame.wordpress.com/2013/11/26/pdo-and-shot-location/, and shows that there is a link between PDO and shot location, and there is a decent correltation for shot locations year on year. 

Some teams, I think Utd being the stand out, have a high PDO year on year. I think there is possibly also a survivor bias in the year on year PDO numbers as I think there is a very good chance the teams with low PDOs are the teams that get relegated. Essentially, I believe that PDO may not regress as much at extreme levels.

Also, I am not suprised that Sunderland's save %, and this their PDO, has dropped massively this season, or that ours has increased. Some of that will be down to luck, some will be down to the different styles played be each team between seasons, but I think a good chunk of it is because we bought their keeper.

[The Woolster]

Also, I'd rather be lucky than unlucky.

[GrkStav]

Hopefully this link will work

http://www.fourfourtwo.com/statszone/8-2013/matches/695025/team-stats/14/OVERALL_04#tabs-wrapper-anchor

Have to say that whilst the number isn't particularly high, going from memory, the events were happening further up the pitch than usual.

Thank you for that.

[The Woolster]

No problem. from there you should be able to check our other matches to compare. That is the Defensive dashboard, but have a play and I'm sure you'll figure it out.

[GrkStav]

No problem. from there you should be able to check our other matches to compare. That is the Defensive dashboard, but have a play and I'm sure you'll figure it out.

It's going to take me a little bit to figure it out but it's definitely a great resource.

I had a look at this: http://www.fourfourtwo.com/statszone/8-2013/matches/695025/team-stats/14/1_PASS_08#tabs-wrapper-anchor

I don't know if that's going to work as a link.It's Passes- Attacking Third.

All the red (unsuccessful) ones, especially from our defensive third and half look awful (and telling) to me. Am I misinterpreting the visual?

Or this? http://www.fourfourtwo.com/statszone/8-2013/matches/695025/team-stats/14/1_PASS_12#tabs-wrapper-anchor

23 out of 41 long passes unsuccessful!

[TSC]

TSR, SOTR, and Fenwich are all limited and have the same issues, they take no account of chance quality, and as we are giving away a lot of shots from distance we will look relatively bad by those metrics. We are actually one of the better teams when it comes to PDO though, The reason for that is that we are creating good quality chances, so have a high scoring rate, and are giving up difficult chances, so have a very high save ratio.

Personally, I'm not a massive fan of PDO being a measure of 'luck' simply because it regresses a lot. The same issue with chance quality that TSR et al I think can still apply, but no ones really looked at it. The first attempt that I have seen was only in the last week or so here http://thebubblegame.wordpress.com/2013/11/26/pdo-and-shot-location/, and shows that there is a link between PDO and shot location, and there is a decent correltation for shot locations year on year. 

Some teams, I think Utd being the stand out, have a high PDO year on year. I think there is possibly also a survivor bias in the year on year PDO numbers as I think there is a very good chance the teams with low PDOs are the teams that get relegated. Essentially, I believe that PDO may not regress as much at extreme levels.

WTF?  It is football and Hull v Liverpool that's being discussed here isn't it? I'd imagine at the end of it all when speaking of the Hull performance a four letter word named 'shit' still better sums up our performance.

[The Woolster]

Our long ball accuracy this season has been 62%, so that is on the low side for us, but it could well be down to the match situation in this case (as well as us simply playing badly), as they would have been happy with just a draw and were able to sit back. If you look at Hull's defensive dashboard, there is a load of clearances in and around the box, prossibly from a fair few of our long ball attempts.

The Iphone/Ipad app is actually better than the website if you have either of those devices.

[The Woolster]

WTF?  It is football and Hull v Liverpool that's being discussed here isn't it? I'd imagine at the end of it all when speaking of the Hull performance a four letter word named 'shit' still better sums up our performance.

There was some discussion about how our underlying performace stats have been poor over the season. I disagreed.

But if you don't follow some of the more advanced football stats, which is quite possibly the majority, then you may well not have a clue what I am on about.

EDIT: I should probably say that the metrics being discussed have been copied from Ice Hockey for football, and the Ice Hockey people have given them really shit names

So yes, it is about football, but not particularly about the Hull game 

[LiverBirdKop]

At times? It seemed every time their centre halves or goalkeeper had the ball Henderson was the further man forward chasing the ball relentlessly. Suarez looked dead on his feet and as usual Gerrard and Lucas struggled on their own away from home.

That's part of the problem. Having one player press high does not make sense at all. All the other team has to do is make an easy pass or two and they've got space to go forward.

Love Henderson's energy, but he still doesn't read the game well and those pressing expeditions leave an already vulnerable midfield even more exposed.

[GrkStav]

Our long ball accuracy this season has been 62%, so that is on the low side for us, but it could well be down to the match situation in this case (as well as us simply playing badly), as they would have been happy with just a draw and were able to sit back. If you look at Hull's defensive dashboard, there is a load of clearances in and around the box, possibly from a fair few of our long ball attempts.

The Iphone/Ipad app is actually better than the website if you have either of those devices.

I don't. I have android-based phone and tablet. Anyway,

If you're a club that generally likes to play the ball from the keeper to the CBs and/or CDM and 'build' attacking play thus, you can expect that clubs may attempt to press you high up the pitch (for them). Experience shows that those clubs that do not have generally been SOL. Therefore, you have to improve your ability to deal with such a high press and/or select the personnel and role-assignments appropriately. You must also select the personnel in such a way that an alternative approach can be used effectively, in case the other club's pressing is particularly skilful and effective and/or your own passing and moving/positioning do not work out as planned. Or, so it seems to me.

If I am right in my conjectures above, our performance vs Hull City was basically the result of failures of execution and player selection/role-assignment.

[JackWard33]

Personally, I'm not a massive fan of PDO being a measure of 'luck' simply because it regresses a lot. The same issue with chance quality that TSR et al I think can still apply, but no ones really looked at it. The first attempt that I have seen was only in the last week or so here http://thebubblegame.wordpress.com/2013/11/26/pdo-and-shot-location/, and shows that there is a link between PDO and shot location, and there is a decent correltation for shot locations year on year. 

Some teams, I think Utd being the stand out, have a high PDO year on year. I think there is possibly also a survivor bias in the year on year PDO numbers as I think there is a very good chance the teams with low PDOs are the teams that get relegated. Essentially, I believe that PDO may not regress as much at extreme levels.

Also, I am not suprised that Sunderland's save %, and this their PDO, has dropped massively this season, or that ours has increased. Some of that will be down to luck, some will be down to the different styles played be each team between seasons, but I think a good chunk of it is because we bought their keeper.

I thought the whole point of PDO was to measure how you are running vs the mean- ie up to now has your team got lucky or unlucky .... Are you in a false position or not ?

Therefore of course it regresses because all PDO measures would be at the mean given enough time
Ours being high will mostly mean we are due to regress ie we've run good up to now

[JackWard33]

This is all true, though as far as I can tell, the more the model takes shot quality into account, the better we look. MCofA's, for instance, which is entirely based on an expected goals model that heavily relies on shot quality, has us as more or less even with the other top 4 hopefuls. The worrying thing is that while predictiveness of offensive shot quality is high, the same isn't really true of shot quality conceded. Anyway, leaving that last piece aside, I think we're actually looking a littler better that your account makes it seem, we're in the top 2 or 3 for shot quality, both offensively and defensively, and a lot of models haven't really incorporated that data yet. I'd say we're top 4 offensively, and mid table defensively. Definitely needs improvement, as you say.

Interesting and I hope you're right
I'm just not sure the offensive side is sustainable as it relies on a very high conversion rate - tougher to sustain now Sturridge is out
Also regardless of quality of shot conceding the volume we are is a problem - it's just very very hard to succeed when you're doing that