Gameweek 3 wasn’t the highest-scoring, with plenty of the big hitters failing to deliver. But that doesn’t mean there wasn’t much to take away from the weekend. There were some interesting developments with Manchester City’s midfield, Arsenal’s defence and Chelsea’s attack, while Hull’s incredible defensive start also deserves a bit more investigation. Here are the main Fantasy Premier League talking points from the weekend’s action. City midfielders looking like a problemIf you were hoping Manchester City’s midfield would provide some reliable FPL options this season, the weekend may have convinced you otherwise. Phil Foden (£7.0m) was benched, while Rayan Cherki (£7.8m) came off for him around the 65th minute, which could indicate that the two will share minutes. Also, with the addition of Iliman Ndiaye (£5.9m) and Enzo Fernandez (£6.9m), this just further packs an already stacked midfield. There is obviously huge upside in owning a City midfielder. The problem is that you need them to actually be on the pitch for those points to arrive. Even if they start, early subs are still going to be a concern, and right now it’s too early to tell who Enzo Maresca favours and who will get the most minutes. This even affects Nico O’Reilly (£6.5m) and Josko Gvardiol (£5.6m), as the former has been playing in midfield and can also compete with Gvardiol at left-back. The situation is worth monitoring right now, which makes City players — bar Erling Haaland (£15.5m) — an avoid right now. I don’t think they are a definite sell if you can bench them this week for another good player, as they do face Sunderland (H) in Gameweek 5. However, if you have Foden or Cherki and can afford a move for Morgan Rogers (£7.6m), then this looks like a solid transfer. Konsa strengthens the Arsenal defensive picture There was also an interesting development in the Arsenal defence. Ezri Konsa (£4.4m) made his first start for Arsenal, while Mikel Arteta explained post-match that Cristhian Mosquera (£5.4m) had “a muscular issue”.That sounds encouraging for Konsa owners and, importantly, should remove some of the immediate concern surrounding his minutes. Konsa was already an extremely attractive option at his price, but the uncertainty over Mosquera means he should at least start the next few games, and there is every chance he keeps starting until William Saliba (£6.0m) is back from injury. There is very little downside to owning an Arsenal defender priced at £4.4m. With Sunderland (A), Leeds (H), Nottingham Forest (A) and Everton (H) during the next five, I would say it’s a great time to buy Konsa now as it doesn’t look like he will stay at that price for long. Chelsea attackers are still the ones to buyChelsea may have lost this London derby, but from an FPL perspective there was still plenty to like. They once again showed they have a strong attack, creating more than a few good chances against the best defence in the league, and there is no doubt they will score plenty of goals. Cole Palmer (£9.6m), Rogers and Joao Pedro (£7.7m) are all extremely appealing options, and a Chelsea triple-up could actually be a very shrewd strategy over the coming weeks. During the next eight games, they come up against Hull (H), Bournemouth (H), Spurs (H), Man United (H), Sunderland (A) and Leeds (H), which all look like great fixtures for attacking purposes. Palmer should be the priority if you don’t own him, as he will not only be the best captaincy option next gameweek but likely in Gameweek 6 as well against Bournemouth. Rogers and Joao Pedro are also great picks in their own right. The former scored the opener in their 2-1 defeat to Arsenal and now has two goals and an assist in three league games.Morgan Rogers celebrates his goal against Arsenal (Adrian Dennis/AFP via Getty Images)A triple-up does carry some risk, particularly if Chelsea struggle for consistency, but for this run of games I think it’s worth the gamble. Can we trust Hull’s defence? Hull have been the surprise package of the season so far. Three clean sheets from their first three games and seven points on the board is an outstanding start. But before we start assuming Hull have suddenly become one of the league’s best defensive teams, I think we need to dig a little deeper. Their expected goals conceded (xGC) sits at 4.86, which is the eighth-worst in the league. They have by no means been the worst defence, but it does clearly suggest they have been very fortunate to have conceded zero goals. At some point, you would expect the results to move closer to the underlying numbers, and I think that time might come next weekend when they travel to Stamford Bridge. They haven’t yet faced an attack of Chelsea’s calibre. Even though I would still be looking to target a Hull defender, as they might still come in handy as your third bench option, it’s about recognising that three clean sheets from three games doesn’t necessarily mean they are going to become a reliable source of defensive points. Sometimes the best decisions come from asking whether a trend is sustainable rather than simply reacting to the results. And in this case, the underlying data suggests that a little caution is warranted.