By Gary Eason
Education editor, BBC News website
The main benchmark in the secondary tables remains the proportion of pupils attaining the Level 2 threshold - five or more GCSEs at grades A* to C or equivalent - including English and maths GCSEs.
On this basis, 107 schools had 100% attainment (compared with 114 last year).
The best - using the average point score as a tie-break - was, as last year, Chelmsford County High School for Girls, a foundation girls' grammar.
The 120 students taking the exams all had at least five good GCSEs and their average points score of 748.2 equates to almost 13 A* grades for each girl.
The worst maintained school was Parklands High School, a non-specialist comprehensive in Speke, Merseyside. Only 1% got Level 2 with English and maths.
But many otherwise excellent independent schools also appear to do badly - because they choose to enter their pupils for International GCSEs (IGCSEs) in English and maths.
Those are not approved for use in state schools - and they are not taken into account in the government's performance tables.
The chief executive of the Independent Schools Council, Jonathan Shephard, said:
"Children who take the demanding IGCSE examinations are given no credit at all. It is time for common sense to prevail and for these silly tables to be scrapped."
But the government defended its approach.
Schools Minister Jim Knight said: "The tables would be absolutely meaningless and completely chaotic if we did include non-accredited qualifications.
"The IGCSE is not compatible with the national curriculum.
"It does not include compulsory study of Shakespeare or any other classic author - which are protected in the national curriculum.
"Nor does the maths IGCSE have a basic non-calculator test."
Among the state sector schools:
- every pupil obtained five good grades including maths and English in 44 of them, compared with 34 last year, out of a total of 3,044 schools with valid results (1.5%)
- less than half the pupils made the grade in 1,818 schools (60%, an improvement on last year's 63%)
- in 639 schools (about one in five) less than 30% of the pupils made the grade - this is the government's new target minimum. Last year it was 789 schools.
There are 3,008 maintained schools with valid English and maths results for both this year and last.
Year-on-year there was no change in 235 schools (7.8%), 1,080 got worse (35.9%) and 1,693 got better (56.3%) - a clear pattern of improvement.
The range of change was from an improvement of 36 percentage points (from 16% to 52%) to a deterioration of 22 points (37% to 15%).
As it happens both the biggest improvement and the biggest deterioration were in Thurrock local authority.
One notable change in the tables this year is the inclusion of a new indicator showing what percentage of pupils attained at least two good GCSEs (or equivalent) in science subjects. Nationally it was 50.3%.
But this also proved controversial.
In schools which offer separate sciences - chemistry, biology and physics - pupils who opted to do only two were not counted, even though the legal entitlement the government has introduced is to do two sciences.
Mr Shephard called the tables "a travesty of fair reporting".
"Children who gain two A* grades in separate science subjects are ranked below children who gain a single grade C in combined science," he said.
But Mr Knight said the government wanted all children to have a thorough grounding in all three sciences, to equip them for further study and life.
"The vast majority of children do double award science which includes separate classes, coursework and exams in physics, chemistry and biology and counts on the new indicator.
"All children now have a legal right to do at least two science GCSEs.
"Highlighting science results shines an important spotlight on every school's performance - as with English and maths.
"The science indicator will focus schools even more on improving learning and teaching - parents should expect no less."
Performance of the 37 academies which feature in this year's results tables varies widely.
At the top, Haberdashers' Aske's Hatcham College in New Cross had 89% of its 204 Key Stage 4 pupils getting level 2 with English and maths (91% last year).
At the other end, it was 7% in The Marlowe Academy in Ramsgate, Kent, and in the John Madjeski Academy in Reading, Berkshire.
In between, 26 had less than 30% of pupils getting Level 2 with English and maths.
The contextual value added (CVA) measure factors in nine different but overlapping elements of children's backgrounds that are known to affect attainment, such as gender, special educational needs and poverty.
What CVA does is predict what a given child's attainment should be based on the actual attainment of other children with similar prior attainment and similar backgrounds.
The top school in the country was Moreton Community School in Wolverhampton, a technology specialist, with a score of 1090.5 on the measure which is based around 1000.
Only 30% of the pupils there had five good GCSEs with English and maths.
At the other end is Shene School in East Sheen, Richmond-upon-Thames, with a CVA of 934.6 and an English-and-maths score of 22%.
Top nationally this year - as last year - was Colchester Royal Grammar School, with 1323.5 average Level 3 points per student (up from 1257.6 in the 2006 exams).
Bottom was George Dixon International School and Sixth Form Centre in Edgbaston, on 253.7.
The overall best performance among local authority schools this year was in Sutton, with 65% getting the English and maths benchmark, average points of 425.5 and a CVA score of 1002.3.
Bottom was Knowsley on 26.5%, average points 336.9 and a CVA of 1003.8.
Last year's worst, Hull, was just above it on 30%, 341.1 points and a CVA 987.1.
On the CVA measure, top was Waltham Forest on 1017.9.
Bottom was Richmond-upon-Thames on 982.5.
At A/AS-level the best was Herefordshire on 863.2 average points per student, the worst was Knowsley again, on 532.0.