Now that Dalglish has been in charge for a reasonable number of games, we can get a more statistically significant estimation of how the team perform under him versus Hodgson. And the result is: no difference. In particular, the model has a term for the "Dalglish boost" (change in performance under Dalglish), and the best fit model has it set to exactly 0.
What do we take away from this? Were the initial results just noise?
I think not. Rather I think the best explanation is that a new manager often gives a temporary boost to the team. We saw this both with Liverpool (where the temporary boost was statistically significant) and probably with Inter as well under Leonardo. Indeed, many pundits have predicted this is what would occur: Liverpool get a morale boost up front, but in the end, it is the overall quality of the players that matters most.
Monday, February 28, 2011
Friday, February 25, 2011
EPL Week 28 Fantasy Team
Based on expected goals scored and conceded, here are the teams to look at this week:
Those grouped together are roughly equivalent. Beyond that, the better choices are listed first.
As usual, these are based on my own predictions for scores. However, as noted before, we can also infer expected goals from the bookies' odds. Their predictions match mine for "defense" and "both". However, for "offense", they rate Man City as the most likely to score. Beyond Man City and Man U, they list all the teams I gave above, but with lower expectations. They also rate Everton as likely to score a reasonable number of goals. As usual, I will pick my team based on my own predictions.
Given that we have fewer choices in defense, it makes sense to start there. The best options from Liverpool are Martin Kelly and Pepe Reina. Then, we can take two City defenders. Noting that Kompany and Richards are both out, I chose Zabaleta and Kolarov. The latter is a nice option given that he will actually play in midfield. Another reasonable choice would be Boateng. Kolo Toure has more points, but Mancini has not been starting him recently.
That leaves one final defender to take from Man U. Vidic and Raphael are the most obvious choices. I picked the former.
Both Newcastle and Aston Villa give good options at strikers in Leon Best and Darren Bent, respectively. Stoke does not offer any great choices here. On the other hand, at midfield, Newcastle and Stoke offer good options in Joey Barton (or Kevin Nolan) and Etherington, respectively, while Aston Villa's options — Ashley Young or Stewart Downing — are not as great. Unfortunately, Etherington is still injured, and it is unclear whether he will play this weekend.
This leaves midfield with the fewest good choices, so we will need to use our final Man U player there. Nani is the obvious choice. The best choices for our last forward are Kevin Doyle or DJ Campbell, of which I picked the latter. With Etherington inured, I fancy Blackpool as even more likely to score. For the final midfield position, in addition to Young and Downing, we should consider Charlie Adam, but he is unfortunately suspended, so I will go with Downing.
This gives us the following 4-3-3 lineup:
As captain, perhaps the safest choice on the face of it would be Nani, especially given that, as a midfielder, he gains points for Man U clean sheets. However, Man U often play very differently away from OT. I will make the unconventional choice and pick Leon Best, as Newcastle have been quite dangerous at home recently, as Arsenal discovered.
Finally, I'll be keeping an eye on Etherington. If he should become available, then I would swap out Charlie Adam for Etherington, whom I would also make captain given that they are facing West Brom and are at home.
UPDATE: Redid the line-up when I noticed that Etherington may not play.
UPDATE 2: Fixed again after discovering that Charlie Adam is suspended.
- Offense: {Newcastle, Stoke, Aston Villa}, {Wolves, Blackpool}
- Defense: Man City, Liverpool
- Both: Man United
Those grouped together are roughly equivalent. Beyond that, the better choices are listed first.
As usual, these are based on my own predictions for scores. However, as noted before, we can also infer expected goals from the bookies' odds. Their predictions match mine for "defense" and "both". However, for "offense", they rate Man City as the most likely to score. Beyond Man City and Man U, they list all the teams I gave above, but with lower expectations. They also rate Everton as likely to score a reasonable number of goals. As usual, I will pick my team based on my own predictions.
Given that we have fewer choices in defense, it makes sense to start there. The best options from Liverpool are Martin Kelly and Pepe Reina. Then, we can take two City defenders. Noting that Kompany and Richards are both out, I chose Zabaleta and Kolarov. The latter is a nice option given that he will actually play in midfield. Another reasonable choice would be Boateng. Kolo Toure has more points, but Mancini has not been starting him recently.
That leaves one final defender to take from Man U. Vidic and Raphael are the most obvious choices. I picked the former.
Both Newcastle and Aston Villa give good options at strikers in Leon Best and Darren Bent, respectively. Stoke does not offer any great choices here. On the other hand, at midfield, Newcastle and Stoke offer good options in Joey Barton (or Kevin Nolan) and Etherington, respectively, while Aston Villa's options — Ashley Young or Stewart Downing — are not as great. Unfortunately, Etherington is still injured, and it is unclear whether he will play this weekend.
This leaves midfield with the fewest good choices, so we will need to use our final Man U player there. Nani is the obvious choice. The best choices for our last forward are Kevin Doyle or DJ Campbell, of which I picked the latter. With Etherington inured, I fancy Blackpool as even more likely to score. For the final midfield position, in addition to Young and Downing, we should consider Charlie Adam, but he is unfortunately suspended, so I will go with Downing.
This gives us the following 4-3-3 lineup:
- Forwards: Best (NEW), Bent (AST), Campbell (BPL)
- Midfield: Nani (MAU), Barton (NEW), Downing (AST)
- Defenders: Kelly (LIV), Zabaleta (MAC), Kolarov (MAC), Vidic (MAU)
- Keeper: Reina (LIV)
As captain, perhaps the safest choice on the face of it would be Nani, especially given that, as a midfielder, he gains points for Man U clean sheets. However, Man U often play very differently away from OT. I will make the unconventional choice and pick Leon Best, as Newcastle have been quite dangerous at home recently, as Arsenal discovered.
Finally, I'll be keeping an eye on Etherington. If he should become available, then I would swap out Charlie Adam for Etherington, whom I would also make captain given that they are facing West Brom and are at home.
UPDATE: Redid the line-up when I noticed that Etherington may not play.
UPDATE 2: Fixed again after discovering that Charlie Adam is suspended.
Thursday, February 17, 2011
A Great Week For Spurs
What a week for Tottenham Hotspur. First, they beat AC Milan, and then the next day, they beat Barcelona! "Wait," you say. "Wasn't Arsenal supposed to play against Barcelona?" That's what I was expecting too, but that is not what happened.
The team that flew to Barcelona clearly did not play possession football. They ceded 66% possession to Barcelona. You might imagine that they tried to play possession football, and only after losing out to Barcelona in this regard, switched to a different style. But you would be wrong. Barcelona's opponent played a different style of football right from the opening whistle.
In particular, Barcelona's opponent played counter-attacking style. Whenever they got possession, they moved up the pitch as quickly as possible, attempting to score against fewer defenders. Furthermore, if you use all 11 players in defense, as Barcelona's opponent did, then you can't simply pass the ball up the pitch. Instead, you need players with pace to run it up pitch. That is, players like Walcott, Nasri, Lennon, Bale, etc.
Indeed, the similarities between Spurs and Arsenal were not limited to using the same basic tactic of counter-attacking. Both teams were similarly defensive.
Both teams defended in a 4-4-2 formation, with the two banks of 4 players close together in order to prevent balls in to the strikers. Neither team sat inside, or even just outside, the penalty box. Both teams pushed their back line of 4 defenders high up the pitch.
When in counter-attack, both teams were reluctant to commit more than 4 players in attack. They kept their entire back line of four as well as two holding midfielders in a defensive position, so that they would be ready when possession was again lost. In some cases, a fullback or holding midfielder would make a run, but once he lost the ball, he would track back into defense.
Both teams also pressed well, even in some cases, high up the pitch. However, the latter was not a huge part of either team's game plan.
Spurs fans must certainly be pleased with both results. Arsenal helped prove that Spurs' style works against the best sides in the world. And since Arsenal's coach and players boast of not having sacrificed their philosophy in playing Barcelona, this shows that Spurs too play "pretty football".
It will be more interesting to see how Arsenal fans respond. From what I have read so far, they seem happy to defend counter-attacking football now. Indeed, some are suggesting they deserve credit for inventing this tactic for beating Barcelona. Not only have they not watched Spurs play, they have not watched Internazionale Milan either, apparently.
The team that flew to Barcelona clearly did not play possession football. They ceded 66% possession to Barcelona. You might imagine that they tried to play possession football, and only after losing out to Barcelona in this regard, switched to a different style. But you would be wrong. Barcelona's opponent played a different style of football right from the opening whistle.
In particular, Barcelona's opponent played counter-attacking style. Whenever they got possession, they moved up the pitch as quickly as possible, attempting to score against fewer defenders. Furthermore, if you use all 11 players in defense, as Barcelona's opponent did, then you can't simply pass the ball up the pitch. Instead, you need players with pace to run it up pitch. That is, players like Walcott, Nasri, Lennon, Bale, etc.
Indeed, the similarities between Spurs and Arsenal were not limited to using the same basic tactic of counter-attacking. Both teams were similarly defensive.
Both teams defended in a 4-4-2 formation, with the two banks of 4 players close together in order to prevent balls in to the strikers. Neither team sat inside, or even just outside, the penalty box. Both teams pushed their back line of 4 defenders high up the pitch.
When in counter-attack, both teams were reluctant to commit more than 4 players in attack. They kept their entire back line of four as well as two holding midfielders in a defensive position, so that they would be ready when possession was again lost. In some cases, a fullback or holding midfielder would make a run, but once he lost the ball, he would track back into defense.
Both teams also pressed well, even in some cases, high up the pitch. However, the latter was not a huge part of either team's game plan.
Spurs fans must certainly be pleased with both results. Arsenal helped prove that Spurs' style works against the best sides in the world. And since Arsenal's coach and players boast of not having sacrificed their philosophy in playing Barcelona, this shows that Spurs too play "pretty football".
It will be more interesting to see how Arsenal fans respond. From what I have read so far, they seem happy to defend counter-attacking football now. Indeed, some are suggesting they deserve credit for inventing this tactic for beating Barcelona. Not only have they not watched Spurs play, they have not watched Internazionale Milan either, apparently.
Wednesday, February 16, 2011
Poor Refereeing in Arsenal v Barcelona
Studies have shown that "home field advantage" is mainly due to the fact that referee decisions are biased in favor of the home team. Probably subconsciously, the ref does not want to upset the crowd and so he biases his decisions toward what will make them happy.
Arsenal v Barcelona was a pretty remarkable example of this. The poor referee decisions that went Arsenal's way included:
Those are four game-changing referee decisions. Most importantly, they were four decisions that, if given differently, would have resulted in a huge uproar from the crowd. Was that part of the reason why the referee made the wrong decisions? Studies suggest it was, at least subconsciously.
Arsenal v Barcelona was a pretty remarkable example of this. The poor referee decisions that went Arsenal's way included:
- Song (Arsenal) received a yellow card for a blatant (tactical) foul of Messi. He continued playing harshly, and shortly thereafter the referee was indicating that he had received his final warning. The referee's hand gestures were: one more and you're gone. Despite that warning, Song committed at least 4 more harsh fouls without being sent off. The worst of the bunch was an attempted two-footed tackle of Messi.
- Messi's goal was disallowed when he was called offside. Careful review shows he was onside, however. I can be somewhat forgiving, here, as it does look questionable when watched at full speed, but nonetheless the decision was incorrect. In the same attack, Arsenal's defense also had a handball in the penalty area that was not penalized.
- Pedro was bowled over by Koscielny in the penalty box. Some thought Pedro was looking for the foul, but you can see in the replay that, after Koscielny bumps into him and Pedro starts to tip, Koscielny puts his arm around him and pulls.
- Arshavin handled the ball in his own penalty box, denying a scoring opportunity. Perhaps the handball was accidental, but given that Arshavin headed the ball into his own hand, that isn't much of an excuse.
Those are four game-changing referee decisions. Most importantly, they were four decisions that, if given differently, would have resulted in a huge uproar from the crowd. Was that part of the reason why the referee made the wrong decisions? Studies suggest it was, at least subconsciously.
Sunday, February 13, 2011
EPL Title Predictions as of Week 27
If we can determine the probabilities for the outcome of every match, it is a fairly simple matter to predict the final results at the end of the season. Treating each match as independent, we can sum up the probabilities of achieving each number of points. Here are the expected values of the resulting distribution:
We can see that there is not much movement from the current standings. In particular, this shows that the games in hand for various teams do not amount to much. This is likely because those most of those matches are between fairly evenly matched teams.
Keep in mind, of course, that all of the numbers above are simply expected values. The associated standard deviations are all between 3.9 and 4.5. So in effect, the race for the Champions League and relegation spots are both very much up in the air.
However, we can say a few things from this. Seeing City pass United would be an event nearly three standard deviations outside the mean, i.e., less than 1 chance in 100. Similarly, Liverpool will not take the Europa league spot. On the other hand, as City and Arsenal are separated by 7.5 points, there is a greater than 1 in 20 chance of City finishing ahead of Arsenal.
- 75.7 — Manchester United
- 72.4 — Arsenal
- 64.9 — Manchester City
- 64.3 — Chelsea
- 63.2 — Tottenham
- 54.5 — Liverpool
- 52.3 — Sunderland
- 50.5 — Bolton
- 50.2 — Newcastle
- 49.3 — Stoke
- 46.7 — Birmingham
- 45.7 — Everton
- 45.4 — Fulham
- 44.7 — Blackburn
- 44.4 — Aston Villa
- 43.4 — Blackpool
- 40.3 — West Brom
- 39.6 — Wolverhampton
- 39.1 — Wigan
- 38.0 — West Ham
We can see that there is not much movement from the current standings. In particular, this shows that the games in hand for various teams do not amount to much. This is likely because those most of those matches are between fairly evenly matched teams.
Keep in mind, of course, that all of the numbers above are simply expected values. The associated standard deviations are all between 3.9 and 4.5. So in effect, the race for the Champions League and relegation spots are both very much up in the air.
However, we can say a few things from this. Seeing City pass United would be an event nearly three standard deviations outside the mean, i.e., less than 1 chance in 100. Similarly, Liverpool will not take the Europa league spot. On the other hand, as City and Arsenal are separated by 7.5 points, there is a greater than 1 in 20 chance of City finishing ahead of Arsenal.
Thursday, February 10, 2011
EPL Week 27 Fantasy Team
Based on expected goals, here are the teams to choose from this week:
While Chelsea gives lots of easy choices for defenders, Blackburn does not. So we need to look for defensive options from Liverpool and/or Manchester United. As it happens, Liverpool does not have any great choices for striker, so this is a natural plan for Liverpool anyway. Manchester United does have good options for striker. However, since there are so many good options from other teams and the fact that Manchester City will be playing very defensively, I choose to follow the same plan both Liverpool and Manchester United: choose one defender and one midfielder. Note that the midfielder also gets points for a clean sheet.
With that plan in place, most of the line-up fills in rather easily, aside from one tricky question: who is the third midfielder. One obvious option is to pick Cesc Fabregas from Arsenal. If we do that, we end up choosing a striker from Blackpool, such as DJ Campbell. On the other hand, we could choose a second striker from Arsenal, namely Theo Walcott. In that case, we would get a midfielder from Aston Villa or Blackpool, of which there are several good options.
I choose the second approach because I feel more confident that Aston Villa will score goals against Blackpool than vice versa. And amongst Aston Villa's midfielders, I picked Ashley Young as his recent form has been slightly better than Stewart Downing's.
The final line-up is the following 4-3-3:
Note: I should also mention that van Persie is currently listed as injured but returning this weekend. If we get news that he will not be available this weekend, then I could either put Fabregas back in or swap van Persie for Arshavin.
- Offense: Arsenal, Aston Villa, Blackpool
- Defense: Chelsea, Blackburn
- Both: Liverpool, Manchester United
While Chelsea gives lots of easy choices for defenders, Blackburn does not. So we need to look for defensive options from Liverpool and/or Manchester United. As it happens, Liverpool does not have any great choices for striker, so this is a natural plan for Liverpool anyway. Manchester United does have good options for striker. However, since there are so many good options from other teams and the fact that Manchester City will be playing very defensively, I choose to follow the same plan both Liverpool and Manchester United: choose one defender and one midfielder. Note that the midfielder also gets points for a clean sheet.
With that plan in place, most of the line-up fills in rather easily, aside from one tricky question: who is the third midfielder. One obvious option is to pick Cesc Fabregas from Arsenal. If we do that, we end up choosing a striker from Blackpool, such as DJ Campbell. On the other hand, we could choose a second striker from Arsenal, namely Theo Walcott. In that case, we would get a midfielder from Aston Villa or Blackpool, of which there are several good options.
I choose the second approach because I feel more confident that Aston Villa will score goals against Blackpool than vice versa. And amongst Aston Villa's midfielders, I picked Ashley Young as his recent form has been slightly better than Stewart Downing's.
The final line-up is the following 4-3-3:
- Strikers: van Persie (ARS), Walcott (ARS), Bent (AST)
- Midfield: Young (AST), Meireles (LIV), Nani (MAU)
- Defense: Samba (BLR), Cole (CHE), Kelly (LIV), Rafael (MAU)
- Keeper: Cech (CHE)
Note: I should also mention that van Persie is currently listed as injured but returning this weekend. If we get news that he will not be available this weekend, then I could either put Fabregas back in or swap van Persie for Arshavin.
EPL Week 27 Bets
I'm doing some major on improving my betting system at the moment. The results suggest that my models are more accurate than the odds compilers' at predicting the number of goals, but less accurate at predicting the outcome of the match. So it seems clear that the mathematics by which I translate between goals and outcomes needs work. I'll report back on that in the future.
Since that is in the works, I'm taking it a bit slower on betting at the moment. However, one bet this week looks too good to pass up:
Even aside from the mathematics, my sense as a football fan is that West Ham has a better chance that the odds compilers are giving them, so this seems like a good bet.
Since that is in the works, I'm taking it a bit slower on betting at the moment. However, one bet this week looks too good to pass up:
- West Ham to win: 27% versus 37%
Even aside from the mathematics, my sense as a football fan is that West Ham has a better chance that the odds compilers are giving them, so this seems like a good bet.
Expected Goals for EPL Week 27
Using the methodology described in this post, here are the expected numbers of goals that we can infer from the bookmakers' odds:
I should point out that my own model has two noticeable differences from this. First, I expect Blackpool to score more goals. Second, I expect Wolverhampton to score more, i.e., an Arsenal clean sheet is less likely in my model.
- 2.46 — Arsenal
- 2.03 — Liverpool
- 1.59 — Manchester United
- 1.57 — West Brom
- 1.50 — Chelsea
- 1.38 — Aston Villa
- 1.33 — Blackburn
- 1.18 — Birmingham
- 1.17 — Tottenham
- 1.15 — Everton
- 1.13 — Bolton
- 1.06 — Blackpool
- 1.06 — Sunderland
- 1.04 — Stoke
- 0.98 — West Ham
- 0.96 — Newcastle
- 0.81 — Manchester City
- 0.75 — Fulham
- 0.54 — Wigan
- 0.48 — Wolverhampton
I should point out that my own model has two noticeable differences from this. First, I expect Blackpool to score more goals. Second, I expect Wolverhampton to score more, i.e., an Arsenal clean sheet is less likely in my model.
Wednesday, February 9, 2011
Steve Clarke on Great Defense
Liverpool's new first team coach, Steve Clarke, tells the club website:
Cheers, Steve!
Your success is always based on how you defend. ... I know it sounds like an old cliché and it’s really simple but the basis of any good team is built on a strong defence.As noted in the last post, Liverpools recent improved form is mostly due to improvements in defense. If various sources are to believed, those defensive improvements are in large part due to Steve Clarke.
Cheers, Steve!
Thursday, February 3, 2011
The Dalglish Boost: Improved Defense
I've been cautiously avoiding making any predictions about Liverpool matches since Roy Hodgson was fired and replaced by Kenny Dalglish. This turns out to have been a wise decision.
My temptation with any change to the team — be it an injury, transfer, or new manager — is to accept the null hypothesis: no change. This is really the only choice since something is changing in almost every match. There are an infinite number of possible variables that could influence the outcome of the match. If we are going to make predictions about what is going to happen, we have to determine which variables are important and which are not. Hence, for many variables, we must accept the null hypothesis, that those variables do not really matter.
It is qualitatively clear at this point, however, that Dalglish's Liverpool is different. They may be the same players (or at least they were until the last match), but they are certainly playing in a distinctly different style. While we don't yet have enough matches to prove that the Dalglish effect is statistically significant, I feel comfortable introducing a new variable at this point, which I will call the "Dalglish boost", the change in Liverpool performance due to Kenny Dalglish.
We have to be careful in introducing the Dalglish boost because we still do not have a lot of data to go by. Dalglish has only been in change for 6 Premier League matches, and he hadn't even practiced with the team for the first of those. We will assume, then, that the Dalglish boost has been in effect for the last 5 Liverpool matches, which is a paltry number.
If we are to take those 5 matches at face value and assume that Dalglish's Liverpool is a whole new team, then the statistics tell us that Dalglish's Liverpool has the best defense in the league and the 4th best offence, slightly ahead of Manchester City in both categories. That seems a bit too optimistic an assumption for only 5 matches.
We can fix this using the usual statistical techniques for controlling overfitting: we introduce a penalty on the complexity of the model. Now, the Dalglish boost will only enter into the equation if it pays for itself: if the increase in model fit is larger than the increase in complexity.
For the purposes of this analysis, we can adjust the penalty by hand until it demonstrates an acceptable level of conservatism. If we set it very high, then Dalglish's Liverpool is no different from Hodgson's. If we set it very low, then Dalgish's Liverpool is better than Manchester City. I've chosen a penalty large enough that Dalglish's team is settled in between Sunderland and Tottenham. However, with virtually any penalty, we see the same qualitative effect, as I will now describe.
This particular model has a term for each team's offensive power and defensive power. Dalglish similarly gets two boosts: one for offense and one for defense. With the penalty above, we get the following results:
This offensive boost is essentially zero. However, a defensive boost of -0.15 is the difference between Sunderland's defense and that of Hodgson's Liverpool. In other words, this model shows a significant Dalglish boost and ascribes it almost wholly to defense.
As I said, this qualitative effect is not particularly sensitive to the penalty chosen above. If you instead decide that Dalglish's Liverpool is now the 4th best team in the Premier League and you choose a penalty that allows the Dalglish boost to get that large, then you get these results:
Now, we see a larger improvement in offense. However, the model still ascribes ~75% of the improvement to defensive improvements.
To those who have watched Liverpool play under Dalglish, this should not be entirely a surprise. Perhaps the biggest visible change instituted by Dalglish has been an increase in pressing. Like the best defense in the world, Dalglish's Liverpool presses whenever they lose the ball, even high up the pitch. This decreases from the pressure on Liverpool's central defense, who (one must admit) were a weak point earlier in the season.
Dalglish has also increased the amount of short passing, thus favoring a more possession-based and less direct approach. This clearly improves defense since the opponent cannot score when you're sitting on the ball.
The end result of these changes is that, while Dalglish's Liverpool have been a more potent attacking force, they have also had 3 clean sheets out of those 5 games. According to the statistics, these clean sheets are a more significant change than the increase in goals. In other words, the primary effect of the Dalglish boost has been improved defense.
My temptation with any change to the team — be it an injury, transfer, or new manager — is to accept the null hypothesis: no change. This is really the only choice since something is changing in almost every match. There are an infinite number of possible variables that could influence the outcome of the match. If we are going to make predictions about what is going to happen, we have to determine which variables are important and which are not. Hence, for many variables, we must accept the null hypothesis, that those variables do not really matter.
It is qualitatively clear at this point, however, that Dalglish's Liverpool is different. They may be the same players (or at least they were until the last match), but they are certainly playing in a distinctly different style. While we don't yet have enough matches to prove that the Dalglish effect is statistically significant, I feel comfortable introducing a new variable at this point, which I will call the "Dalglish boost", the change in Liverpool performance due to Kenny Dalglish.
We have to be careful in introducing the Dalglish boost because we still do not have a lot of data to go by. Dalglish has only been in change for 6 Premier League matches, and he hadn't even practiced with the team for the first of those. We will assume, then, that the Dalglish boost has been in effect for the last 5 Liverpool matches, which is a paltry number.
If we are to take those 5 matches at face value and assume that Dalglish's Liverpool is a whole new team, then the statistics tell us that Dalglish's Liverpool has the best defense in the league and the 4th best offence, slightly ahead of Manchester City in both categories. That seems a bit too optimistic an assumption for only 5 matches.
We can fix this using the usual statistical techniques for controlling overfitting: we introduce a penalty on the complexity of the model. Now, the Dalglish boost will only enter into the equation if it pays for itself: if the increase in model fit is larger than the increase in complexity.
For the purposes of this analysis, we can adjust the penalty by hand until it demonstrates an acceptable level of conservatism. If we set it very high, then Dalglish's Liverpool is no different from Hodgson's. If we set it very low, then Dalgish's Liverpool is better than Manchester City. I've chosen a penalty large enough that Dalglish's team is settled in between Sunderland and Tottenham. However, with virtually any penalty, we see the same qualitative effect, as I will now describe.
This particular model has a term for each team's offensive power and defensive power. Dalglish similarly gets two boosts: one for offense and one for defense. With the penalty above, we get the following results:
- Dalglish offensive boost: 0.02 goals per game (scored)
- Dalglish defensive boost: -0.15 goals per game (conceded)
This offensive boost is essentially zero. However, a defensive boost of -0.15 is the difference between Sunderland's defense and that of Hodgson's Liverpool. In other words, this model shows a significant Dalglish boost and ascribes it almost wholly to defense.
As I said, this qualitative effect is not particularly sensitive to the penalty chosen above. If you instead decide that Dalglish's Liverpool is now the 4th best team in the Premier League and you choose a penalty that allows the Dalglish boost to get that large, then you get these results:
- Dalglish offensive boost: 0.12 goals per game (scored)
- Dalglish defensive boost: -0.32 goals per game (conceded)
Now, we see a larger improvement in offense. However, the model still ascribes ~75% of the improvement to defensive improvements.
To those who have watched Liverpool play under Dalglish, this should not be entirely a surprise. Perhaps the biggest visible change instituted by Dalglish has been an increase in pressing. Like the best defense in the world, Dalglish's Liverpool presses whenever they lose the ball, even high up the pitch. This decreases from the pressure on Liverpool's central defense, who (one must admit) were a weak point earlier in the season.
Dalglish has also increased the amount of short passing, thus favoring a more possession-based and less direct approach. This clearly improves defense since the opponent cannot score when you're sitting on the ball.
The end result of these changes is that, while Dalglish's Liverpool have been a more potent attacking force, they have also had 3 clean sheets out of those 5 games. According to the statistics, these clean sheets are a more significant change than the increase in goals. In other words, the primary effect of the Dalglish boost has been improved defense.
EPL Week 26 Fantasy Team
Here's what my own models say about this weekend's matches.
The best purely offensive choices are Everton and Arsenal. Both of these teams are playing against opponents (Blackpool and Newcastle) that are also expected to score goals, so these are not safe defensive options. Everton has only one obvious choice: Tim Cahill as striker. Arsenal's midfield often has too many good choices to make any of them safe. However, with Sami Nasri currently injured, Cesc Fabregas becomes an easy choice.
The best purely defensive choices are Aston Villa, Fulham, Tottenham, Bolton, Stoke, and Sunderland. Aston Vill is the most likely clean sheet, but they also have the fewest good choices. Fulham has many good choices for defender, but it will be harder for them to keep a clean sheet against a revitalized Aston Villa. I decided to hedge my bets by picking one defender from each of the italicized teams.
Finally, the teams that look the best both offensively and defensively are Man United, Man City, and Chelsea. Man United is a bit stronger offensively than defensively, while Man City and Chelsea are the reverse, according to the model. However, given Liverpool's recent run of form and new signings, I feel less confident about Chelsea's defense. One choice is easy however: since none of the defensive teams mentioned above have good options for goalie, I picked Joe Hart from Man City.
After choosing Tim Cahill, we are left with 2 strikers and 3 midfielders to choose from amongst Arsenal, Man United, Chelsea, and Man City.
Perhaps the safest bet this week for striker is Wayne Rooney. Even when he does not score goals, he gets assists, which are valuable from a striker. And of course, he found the net twice in the last game. For the other choice, I am split between Tevez and Torres, with the latter seeming likely to find the net on his Chelsea debut. This is pretty much a toss-up as each of them have potent strike partners that could also pick up goals. Ultimately, though, Tevez seems the safest bet because he too gets assists.
In midfield, the obvious choice from Man United is Nani. Park Ji Sung is also a reasonable choice, but it's hard to know if he will be in the starting lineup. From Chelsea, the best options are Malouda and Lampard. Choosing between those two is difficult, especially given that Lampard is Chelsea's penalty taker.
One option would be to choose both Malouda and Lampard. It is likely that at least one of them will be involved in a goal. Furthermore, that uses only one player from Manchester United, so I could then go back and add Vidic as a defender.
However, Liverpool's defense has improved considerably, so betting on Chelsea goals does not seem safe, while Man United is likely to ring up a high score against Wolverhampton. So instead, I'll pick Nani and, after flipping a coin, Malouda. This does leave the option of taking a Chelsea defender, but again I worry about the new form of Liverpool.
Here is the final lineup:
The final result is fairly standard in offense, despite all that analysis, but the defense is perhaps surprising.
The best purely offensive choices are Everton and Arsenal. Both of these teams are playing against opponents (Blackpool and Newcastle) that are also expected to score goals, so these are not safe defensive options. Everton has only one obvious choice: Tim Cahill as striker. Arsenal's midfield often has too many good choices to make any of them safe. However, with Sami Nasri currently injured, Cesc Fabregas becomes an easy choice.
The best purely defensive choices are Aston Villa, Fulham, Tottenham, Bolton, Stoke, and Sunderland. Aston Vill is the most likely clean sheet, but they also have the fewest good choices. Fulham has many good choices for defender, but it will be harder for them to keep a clean sheet against a revitalized Aston Villa. I decided to hedge my bets by picking one defender from each of the italicized teams.
Finally, the teams that look the best both offensively and defensively are Man United, Man City, and Chelsea. Man United is a bit stronger offensively than defensively, while Man City and Chelsea are the reverse, according to the model. However, given Liverpool's recent run of form and new signings, I feel less confident about Chelsea's defense. One choice is easy however: since none of the defensive teams mentioned above have good options for goalie, I picked Joe Hart from Man City.
After choosing Tim Cahill, we are left with 2 strikers and 3 midfielders to choose from amongst Arsenal, Man United, Chelsea, and Man City.
Perhaps the safest bet this week for striker is Wayne Rooney. Even when he does not score goals, he gets assists, which are valuable from a striker. And of course, he found the net twice in the last game. For the other choice, I am split between Tevez and Torres, with the latter seeming likely to find the net on his Chelsea debut. This is pretty much a toss-up as each of them have potent strike partners that could also pick up goals. Ultimately, though, Tevez seems the safest bet because he too gets assists.
In midfield, the obvious choice from Man United is Nani. Park Ji Sung is also a reasonable choice, but it's hard to know if he will be in the starting lineup. From Chelsea, the best options are Malouda and Lampard. Choosing between those two is difficult, especially given that Lampard is Chelsea's penalty taker.
One option would be to choose both Malouda and Lampard. It is likely that at least one of them will be involved in a goal. Furthermore, that uses only one player from Manchester United, so I could then go back and add Vidic as a defender.
However, Liverpool's defense has improved considerably, so betting on Chelsea goals does not seem safe, while Man United is likely to ring up a high score against Wolverhampton. So instead, I'll pick Nani and, after flipping a coin, Malouda. This does leave the option of taking a Chelsea defender, but again I worry about the new form of Liverpool.
Here is the final lineup:
- Goalkeeper: Hart (MAC)
- Defenders: Collins (AST), Hangeland (FUL), Huth (STO), Bardsley (SUN)
- Midfielders: Fabregas (ARS), Malouda (CHE), Nani (MAU)
- Forwards: Cahill (EVE), Tevez (MAC), Rooney (MAU)
The final result is fairly standard in offense, despite all that analysis, but the defense is perhaps surprising.
Expected Goals for EPL Week 26
Using the same methodology as discussed in a previous post, here is what we can infer from the odds about the expected number of goals for each team in week 26:
- 1.76 — Everton
- 1.73 — Manchester United
- 1.70 — Tottenham
- 1.69 — Manchester City
- 1.67 — Chelsea
- 1.53 — Arsenal
- 1.42 — Aston Villa
- 1.39 — West Ham
- 1.22 — Stoke
- 1.19 — Wigan
- 1.03 — Birmingham
- 1.02 — Blackburn
- 0.96 — Sunderland
- 0.86 — Fulham
- 0.77 — Newcastle
- 0.72 — Liverpool
- 0.64 — Blackpool
- 0.59 — Bolton
- 0.53 — Wolverhampton
- 0.49 — West Brom
Subscribe to:
Posts (Atom)