Gameweek 3: Where the Points Are

The deadline is 15:25 in Lisbon on Saturday 22 August, five minutes before Marítimo play Académico de Viseu.

Seven games, not nine. Benfica, Braga, Moreirense and Estrela Amadora do not play, and there is a short note about them at the end. Everything before that is the seven games that do happen, in the order they kick off, and who we expect to score in each.

Two of the numbers below need a word first. Expected goals is what a team's chances are worth, so 2.50 means two and a half goals of chances created, not 2.50 goals scored. Per million is a player's expected points divided by his price, and the league average this week is 0.61: above it he pays for himself, below it he does not.

Marítimo v Académico de Viseu

Saturday 15:30, Barreiros. Marítimo are 49% to win, the draw is 26% and Académico are 25%. We expect Marítimo to create 1.43 goals of chances against 0.93, and we give them a 39% chance of a clean sheet against Académico's 24%.

Both clubs came up this summer, and one of them has six points from six. Marítimo beat Casa Pia 1-0 and then won 2-1 at Famalicão. Nobody has noticed: the entire Marítimo squad carries less ownership than Diogo Costa does on his own.

This is the cheapest good attack of the week, and it is not close. The four highest projections in this game are all Marítimo players, and not one of them costs more than €5.5.

Butzke at €5.5 is the best value forward of the week at 0.88 per million, and 3.3% of managers own him. Rafael Fernandes at €4.5 returns more per million than anybody in the league this week, at 0.93, though he carries our lowest confidence band: he is a defender at a promoted club with two league games of record behind him.

Marítimo are without Nathanael Ogbeta, Madsen left the island this week, and Maxime Dominguez is back in contention after a muscle problem. Académico have three doubts in Gustavo Costa, Tomás Silva and Luís Silva (zerozero).

From the away side, only André Clóvis is worth a look, at €6.0 and 17.5% owned. He is the penalty taker and the whole attack, and we expect 3.74 points. His club is 25% to win.

Estoril Praia v Rio Ave

Saturday 18:00, Coimbra da Mota. Estoril are 51% to win, the draw is 23% and Rio Ave are 25%. We expect 1.62 goals of chances from Estoril and 1.05 from Rio Ave, with clean sheet chances of 35% and 20%.

One point and zero points. Rio Ave have not scored a league goal yet. Estoril lost 2-0 at Nacional with Xeka sent off after half an hour, and he is suspended this week.

This is the most open game of the week. We give it a 52% chance that both teams score, the highest of the seven, which is what makes the clean sheet numbers here worse than the win numbers suggest.

Begraoui at €7.5 is the obvious pick and the right one. He takes the penalties, he scored the equaliser against Famalicão, and we expect 5.27 points. He is owned by 38.2%, so he wins you nothing against the field, but he loses you nothing either.

The cheaper route is the Estoril defence, which nobody owns. We expect 4.02 points from Ricard Sánchez at €5.0, owned by 0.9%, which is 0.80 per million. From Medrano at €4.5 we expect 3.88 points, and he carries our low confidence flag.

Rio Ave have one name above 3.7: Diogo Bezerra at €6.0, expected 3.90 points and owned by 0.6%. That is a lot of faith in a club that has not scored yet.

Sporting CP v FC Alverca

Saturday 20:30, Alvalade. Sporting are 81% to win, the draw is 12% and Alverca are 7%. We expect Sporting to create 2.82 goals of chances against 0.73, which is the widest gap of the week.

The biggest attacking projection of the week, against the worst defence of the week. We give Alverca a 6% chance of a clean sheet, the lowest of the fourteen clubs playing.

M. Araújo is the defender we expect most from this week, at 5.45 points.

Doumbia at €7.5 looks like the value pick, and he comes with a warning. We give him 2.23 points from goals alone, more than any other midfielder in the league this week, and he costs €3.1 less than Zalazar. It is worth knowing where that number comes from.

Doumbia was the star of Sporting's pre-season. Two goals and an assist in five starts, and Rui Borges tried him on the left, at ten and at eight, with the goal and the assist against Nottingham Forest coming from an advanced role (JN, Record). Behind that sits a real scoring record: 8 goals from 70 shots in 37 Serie B games for Venezia last season. Our model reads all of it.

In the league he has been playing much deeper. He has started both games in the two-man midfield beside Altimira, played almost every minute of each, and in that time he has taken zero shots and made five tackles. A holding midfielder who does not shoot does not return 2.23 points of goals. If Rui Borges keeps him there, the 6.29 beside his name is too high, and the safe version of this fixture is Zalazar or Araújo. If he pushes forward again, Doumbia is the best value on the board. This is the one projection this week we would not defend hard.

Rui Silva is the quiet one. Sporting have the second best clean sheet chance of the week at 48%, and their goalkeeper costs €6.0 and is owned by 7.3%. We expect 4.39 points from Porto's Diogo Costa against his 4.27, and he costs €0.5 more and is owned by 60.7%. The two goalkeepers are just over a tenth of a point apart.

The number nine question has been answered. Ioannidis has trained apart all week with lower back pain and is a doubt. Rui Borges said he would decide after Friday's session, and Luis Suárez is the man in line. He has 53 minutes off the bench so far, and the Portuguese press expect this to be his first start of the season (Record). It is settled now, and our numbers moved with it: Suárez starts, and at 7.02 he is the highest projection in the league this week, while Ioannidis is 0.47. If you own neither, none of it matters: Zalazar, Doumbia, Araújo and Rui Silva are unaffected either way.

Alverca have Chiquinho suspended and nothing worth buying. The most we expect from anybody there is 3.18 points, from Figueiredo.

Vitória SC v CD Nacional

Sunday 15:30, D. Afonso Henriques. Vitória are 56% to win, the draw is 23% and Nacional are 21%. We expect 1.65 goals of chances from Vitória and 0.88 from Nacional, with clean sheet chances of 42% and 19%.

Vitória have no points and have played well in both defeats. They created 3.34 expected goals and allowed 1.61 in them, which is the third best expected goal difference in the league behind only Benfica and Porto. They are sixteenth.

Samu at €6.5 is our strongest recommendation of the week. He takes the penalties, he scored against Sporting, and we expect 5.35 points. Only 14.2% own him. He is the best midfielder under €7.0 on the board.

Ndoye scored against Sporting as well, and costs the same as Samu. Balieiro at €4.5 is the cheap way into a defence we give a 42% clean sheet chance, the fourth best of the week.

Tiago Margarido is still waiting for his first league point at Vitória, and he gets his old club to try it against: he coached Nacional for three seasons before moving to Guimarães in June (Maisfutebol).

Nacional arrive with four points and a defender worth watching rather than buying: Zé Vítor at €5.0 has 19 points already, more than any other defender in this fixture, but we expect 3.33 points from him away from home here.

One warning about this game. Our model still gives Chucho Ramírez 4.17 points for Nacional. He signed for Çorum FK on 7 August (O Jogo). We work out minutes from a playing record rather than from the transfer flag, so that number is simply wrong. Ignore it.

Santa Clara v FC Famalicão

Sunday 18:00, São Miguel. Santa Clara are 38% to win, the draw is 27% and Famalicão are 36%. We expect 1.18 goals of chances from Santa Clara and 1.14 from Famalicão, with clean sheet chances of 32% and 31%. Nothing separates them.

The closest game of the week and the quietest: 2.32 expected goals in total, against 3.55 at Alvalade. Two clubs that cannot be separated, in a game that may not produce much.

It also carries more ownership than any game that does not involve Porto or Sporting. Three of the five most owned players here return less than the average.

• Rodrigo Pinheiro at €5.5 is owned by 25.5%. He is the second most owned defender in Liga Portugal, behind only Kiwior. We expect 3.25 points, or 0.59 per million.

• Gonçalo Paciência at €5.7 is owned by 27.6% and has three goals, second only to Pavlidis. We expect 3.42 points, or 0.60.

• De Amorim at €5.5 is owned by 16.9% for 3.17, which is 0.58.

Two of the five do clear it. Carevic at €5.0 returns 0.71 behind a defence we give a 32% clean sheet chance, and Famalicão's Sorriso at €7.0 is the most we expect from anybody in the game, at 4.49 points. Santa Clara have a Sorriso too. They are different men, and they play against each other.

If you hold two or three of these, this is the game to think hardest about, because the ownership was earned in two games and the fixture is the quietest of the seven.

FC Porto v FC Arouca

Sunday 20:30, Dragão. Porto are 80% to win, the draw is 13% and Arouca are 7%. We expect Porto to create 2.55 goals of chances against 0.58, the smallest attacking projection given to any club this week.

Porto have the best clean sheet chance of the week at 56%. Arouca have the second worst at 8%.

Arouca are top of the league. Most likely not for long.

They have won both games, scored five and conceded none. In those same two games they created 2.63 expected goals and allowed 2.71. Their real goal difference is plus five and their expected goal difference is minus 0.08, the biggest gap in the league. Now they go to a club that has conceded nothing in two games.

Three Arouca players are owned by more than one manager in ten, and all three are under the average this week: Fontán at 1.82, Barbero at 2.69 and Djouahra at 2.63.

On the Porto side, the name at the top of the board is the one hardly anybody has.

André Silva at €10.0 is the pick of this game, owned by 9.0%. He is the listed penalty taker, we expect him to start, and we expect 6.45 points: the most of any Porto player this week, and 1.3 clear of the next one.

Nehuen Pérez at €5.5 returns more per million than any other Porto player, and 2.1% of managers own him. He has 14 points from 85 minutes. Behind him, Gabri Veiga at €8.1 and Froholdt at €8.5 are four hundredths of a point apart, and Froholdt is owned by 44.5% against Gabri Veiga's 12.7%. Anyway we don't know for how long he will stay on porto's starting XI.

Porto have seven players receiving treatment. Martim Fernandes is out, and Alan Varela and Oskar Pietuszewski are close to returning but not for this one. Samu is out with a known ruptured cruciate ligament (O Jogo).

Gil Vicente FC v Casa Pia AC

Monday 20:15, Barcelos. Gil Vicente are 59% to win, the draw is 22% and Casa Pia are 18%. We expect 1.72 goals of chances from Gil Vicente and 0.82 from Casa Pia, with clean sheet chances of 44% and 18%.

This is the best cheap game of the week, and it is the last one, on Monday night. Your team is locked from Saturday afternoon, so these picks have to be made two days before they play.

Casa Pia have no goals, eight conceded, 0.76 expected goals created and 4.32 allowed in two games. Each of those four is the worst or joint worst in the league. The scoreboard and the underlying numbers agree about them, which is more than can be said for Arouca.

Gil Vicente have the third best clean sheet chance of the week at 44%, behind only Porto and Sporting, and they are far cheaper than either.

Lucão at €4.5 is the cheapest goalkeeper worth owning this week. Four clubs have better than a 40% chance of a clean sheet, and the other three first choice goalkeepers cost €5.0, €6.0 and €6.5.

Santi García is the pick, with one thing to know. The game lists him as García. He is the most we expect from anybody in this fixture, at 4.82 points, and he returns 0.88 per million. Anderlecht and Toulouse both want him, Gil Vicente will only sell for around €12 million, and neither club has come close to that (O Jogo). He is expected to play on Monday. He is also the kind of player who can be gone before the window shuts, so buy him for this game rather than for October.

Murilo is the name to watch, not to buy. He costs €6.4 and 7.2% of managers already hold him, but he has not played a minute this season with a muscle injury. Record reported on 19 August that he is in the last stage of recovery and could make the squad for Monday (Record). We give him an 8% chance of starting and 1.10 points this week, then 2.94 in gameweek 4 and 4.40 in gameweek 5 as he works back in. He had 11 goals and 4 assists in 33 games last season. Weverson is also close after a shoulder injury; Mohamed Bamba and Walace França are further away.

Casa Pia have one name over 3.5: Henrique Araújo at €5.5, expected 3.69 points, on our low confidence flag.

The captain

Suárez is the only player above seven, and he is there because Sporting's number nine question resolved his way. Zalazar behind him takes the penalties in a game Sporting are 81% to win, and is owned by 47.8%, so captaining him wins you very little on the field even when it comes off.

André Silva doubled is 12.90 against Zalazar's 13.42. That is half a point between the safe armband and one that nine managers in ten do not own.

The popular names our numbers do not back

Every player here is owned by more than one manager in ten and returns less than the 0.61 average.

Suárez is the odd one here. He is the highest projection in the league this week and still under the average, because at €12.4 he has to be: he needs a return nobody else on this table is being asked for. The rest is a pattern: Arouca's start and Famalicão's ownership are both built on two games.

The four clubs with no game

Benfica, Braga, Moreirense and Estrela Amadora had their games moved so Benfica and Braga could play European qualifiers. Benfica now play at Moreirense on 9 September and Braga at Estrela Amadora on 10 September, which means all four play twice in gameweek 5.

Players from those four clubs make up 23.8% of everything managers own, so the average squad has 3.56 players scoring nothing this week.

Over the three gameweeks that start this weekend, Pavlidis is still the seventh highest scorer on the board with a zero in the first of them. Abraham Marcus at €6.0 is the one nobody has: 4.5% ownership, and 13.91 across the three, which is nine tenths of a point behind Pavlidis at half the price.

Where these numbers come from

The club numbers come from a goal model built on the quality of the chances each team creates and allows, then adjusted to the betting market, which has priced all seven games this week. The player numbers are built from those, line by line, under Liga Portugal's own scoring rules.

We read everything here at 03:30 in Lisbon on 22 August, twelve hours before the deadline. The projections update through the day as team sheets land.

Two honest limits. Predicting who starts is the hardest part of this, and almost every disagreement above is about minutes rather than about talent. And two games is a very small sample, so the expected goals from gameweeks 1 and 2 describe what has happened rather than forecast what will.