FPL Broccoli

Earlier in the season I wrote a piece on Free Hit risk and Correlated Variance (CoVar) that gained quite a bit of traction. I’m on Free Hit again this week, so I’ve revisited it with some added thoughts and a focus on this gameweek. If you haven’t read the first one, I’ll link it below as it’s worth checking out first. New topics: • Field Correlated Variance (CoVar) • Expected value vs variance trade-off • Applying CoVar to Free Hit GW34 Field Correlated Variance (CoVar) CoVar only really works when we consider the field, meaning what everyone else is picking. On a Free Hit, we can get a good read on ownership for that specific gameweek and identify which players are likely to be highly owned across both Free Hit and non-Free Hit teams. That gives us a short-term ownership picture we can use to make better decisions. If our aim is to maximise upside, we should look to differentiate from these popular picks to some extent (more on this later). This reduces our exposure to highly owned players whose returns offer limited rank gains. Beyond that, we can also actively target against the field. For example, if a defence is expected to be highly owned, we could instead move to a lower-owned attacker from the opposing team with strong projected points (EV), creating a CoVar play against the field. This can be pushed further by doubling up on that attack to increase upside. Expected value vs variance trade-off Following on from this, it’s not as simple as just picking low EO, high EV players against the field. The field is generally trying to maximise EV by targeting strong defenders against weak attacks and strong attackers against weak defences, which makes going against it difficult. So should we sacrifice some EV for higher upside? In most cases, standard CoVar stacking can be achieved without sacrificing much EV. However, with Field CoVar, more EV typically has to be sacrificed, but the upside is also greater. The amount you trade off depends on your risk appetite, your current rank, and how comfortable you are with downside. That said, some players simply have too much EV to ignore and remain essential, regardless of ownership. For context, I’m currently ranked 415th. My season priority was to finish inside the top 10k, then push for top 1k and then top 100. It’s about the digits for me, I value finishing 9,999 the same as 1,001. Because of that, I’m willing to take on more CoVar/Risk with this Free Hit than I would if I felt at risk of dropping outside the top 10k. Applying CoVar to Free Hit GW34 Best types of CoVar/risk in order: 1. Low ownership captaincy with little or no EV sacrifice 2. Multiple defenders from the same team 3. Avoiding opposing attackers and defenders 4. Combining attackers and defenders from the same team (especially set-piece takers with set-piece threats) As discussed, these approaches are most effective when ownership is low. This isn’t as CoVar-friendly a week as I’d like. With Gabriel, Salah, Bruno, Bowen and possibly Virgil looking like locks, it’s difficult to build CoVar around them without sacrificing a chunk of EV. For example, tripling up on Liverpool’s defence feels suboptimal given how strong their midfield options project this week. The same applies to Arsenal, where beyond Gabriel and Raya, Saliba doesn’t project particularly well. I’m even slightly tempted by a Spurs triple defensive stack, as crazy as that sounds, but Solanke also projects well, which works against it. CoVar/Risk I like on Free Hit GW34: • Salah Captain • Porro + Danso + Kinsky • Xavi + Solanke • Raya + Gabriel + Saliba • Tarkowsk + Keane • Triple Liverpool attack Thank you for reading! Special mention to @Spoish and @AnalystNik11, who I spoke to while putting this together. Both are well worth a follow. @AnalystNik11 even has an FPL model, with a CoVar setting available on the solver.