FPL Gameweek 6 Midweek Update: Transfers, Flags and the Model

This is a follow up to Monday's preview, not a replacement for it. The deadline is still Saturday 10 October at 11:00 UK time, and the numbers below are from the model as of Wednesday lunchtime.

What has changed since Monday

Almost nothing. João Pedro, Palmer, Isak, Semenyo and Gakpo are all still on 75%, and their projections are the same to two decimals. The top three are still Saka (5.58), Haaland (5.47) and Bruno Fernandes (5.42). Everything I wrote about the captaincy on Monday stands.

The one change is a small one. Saka has gone up to £9.6m and is now in 14.2% of teams. If you were planning to bring him in, that is 0.1 you no longer have.

The selling has carried on anyway. João Pedro is up to 363,000 out, from 342,000 on Monday, with no change to his flag. Bruno is at 163,000 out. I said on Monday I did not see why, and I still do not.

The buys: three of the top four are Brighton

Player	Net in	Opponent	GW6	Model rank
Groß (BHA, £5.9m)	864,000	Sunderland (a)	4.68	14
Schade (BRE, £6.2m)	419,000	Aston Villa (a)	4.73	12
Kostoulas (BHA, £5.6m)	340,000	Sunderland (a)	3.84	69
De Cuyper (BHA, £5.0m)	314,000	Sunderland (a)	3.53	96
Hall (NEW, £5.3m)	259,000	Coventry (a)	4.78	10
Tarkowski (EVE, £6.2m)	236,000	Hull (a)	5.24	4

About 1.5 million net transfers have gone into three Brighton players. I understand it. Groß is the top scorer in the game, Kostoulas has two 10 pointers in a row, and De Cuyper has 38 points as a £5.0m defender. That is what a good five weeks looks like.

The model is less sure, and the reason is the calendar. Brighton are away at Sunderland this week and at home to Palace in GW7. Then they go to Anfield and to the Etihad. De Cuyper is at 3.53 this week and 4.63 against Palace, then 2.89 and 2.39. Kostoulas goes 3.84, 4.21, 3.67, 3.63. Groß holds up better because his points do not depend on a clean sheet, but he drops too, from 5.29 against Palace to 4.38 and 4.27.

So if you bought a Brighton player this week, you bought two decent fixtures and then two bad ones. That can still be fine. Just know that you will probably be looking at the same slot again in a fortnight, and two Brighton players is one more than I would want going into GW8.

The other three are the opposite story. Schade, Hall and Tarkowski are all in the model's top twelve for the week, so here the market and the model agree. Hall is the one I like most over the run. He takes corners and free kicks, and Newcastle's next five are Coventry, Villa, Palace, Everton and Fulham.

The sells: Szoboszlai is the odd one out

Most of the selling has a reason behind it. João Pedro, Palmer and Tzolis are flagged, and Elanga is out with no return date. Selling a flagged player on a Wednesday is early, as I keep saying, but at least there is a reason.

Szoboszlai is different. He is the second most sold player of the week, 204,000 out, and he is fully fit. He has played 440 of 450 minutes and is on penalties, free kicks and corners. The problem is 20 points from five games, and a price that has already dropped to £6.9m.

He was in the running cold table in the piece from last week, one goal involvement from 2.24 expected. The model has him at 4.16 at home to City, 36th overall. That is not a number to buy for. It is also not the number of a player who has stopped doing anything. If you are selling him to fix a real problem, go ahead. If he is the transfer because he is the most boring name in your midfield, I would leave it.

Arsenal have five flags, and Saka is not one of them

Player	Owned	Flag	GW6
Konsa (£4.6m)	19.1%	75%, unspecified	2.68
Tzolis (£6.3m)	11.8%	75%, hamstring	2.83
Rice (£7.4m)	11.4%	75%, unspecified	3.09
Havertz (£7.6m)	8.0%	75%, hamstring	3.27
White (£5.5m)	4.8%	75%, groin	3.47

I left these out on Monday because none of them is a captain option, and that was a mistake. Arsenal at home to Leeds is the fixture that puts Saka top of the week, and five of his team mates are flagged for it.

Konsa is the one that matters for most squads. He is in nearly one team in five, at £4.6m. He has played 281 of a possible 450 minutes, and the model had him as a rotation risk before the flag. At 2.68 he should not be in your eleven this week unless you have nothing else.

None of this changes Saka's number. He is not flagged and the model has him down for the full 90. But if you own two or three from this table, the Emirates is where your Saturday morning problem is, not Stamford Bridge.

One name from further down: Emersonn

The sixth best projection of the week belongs to an Ipswich forward in 3.7% of teams. Emersonn is at 5.01, at home to Fulham, which FPL rates as a 2. Only Haaland is higher among forwards. He costs £5.5m and he is 93% of the way to a price rise on the prices page, so he may well be £5.6m before the deadline.

Here is the part to be careful with. He has two goals and an assist from 317 minutes, and his scores so far are 9, 1, 2, 12 and 2. That is two good games and three blanks, on a small sample, and the model is treating him as a 90 minute player when he has averaged 63. Then it is City away in GW7, where he drops to 4.24.

So I am not telling you to buy him. I am saying that if you are selling João Pedro this week and want the money elsewhere, he is a better one week answer than Kostoulas at about the same price, and the model says so by more than a point.

Nothing has changed at the top. The market spent the break buying Brighton, and the model thinks that is a two week trade. Check your Arsenal players before you check anyone else.

What I would do

Same as Monday: nothing until the press conferences. If you have Konsa in your eleven, find out whether he trains before you rely on him. If you already bought Groß, keep him and do not add a second Brighton player. And if Szoboszlai is the transfer you were about to make, look at what the model says about who you are buying first.

You can check all of this against your own squad on My Team, and the fixture run for any club is on the fixtures page.