Chelsea picks are this week’s flavour. I compiled the most-asked questions and shared my opinion. Short, sweet, and straight to the point.
Let’s go.
1️⃣ Are Hull good defence?
Look, you need to be careful adopting the opinions of people who just discovered football analytics exists and use the word regressionbutwithnocontext to describe pretty much everything that happens early in the season.
But let’s not pretend Hull can keep defying the laws of probability. They managed it in the previous campaign*, when they conceded 61 actual goals from 75 predicted, and got 72 points from… 54 predicted (oof). However, those 61 goals were still the 6th most in the Championship.
*There was nothing that would stand out in the 21/22, 22/23, and 23/24 seasons. In the 24/25? Pretty unlucky: 13th in xPoints, 21st in reality. So, yeah…
It’s a different team in a different context now, but I wanted to show that even massive overperformance doesn’t mean they won’t start conceding.
Some say they concede a high volume of low-quality shots from distance, wide angles, and blocked shots because so many bodies are between the attacker and their goalkeeper. Their low-block is perfect!
Uhm, their low-block is… well, really low. It’s the lowest out there, with an average distance of defensive touches of about 41 metres from their goal and a 29% possession percentage. Both are extremely low. But their 0.11 xG/shot conceded isn’t out of the ordinary (8 other teams are on par or worse), and if you look at those shots, it’s not as if all of them are poor:
44 shots conceded, but just 12 out of the box
24 centrally
And when you dig even deeper, you find that the opponent’s players posted terrible xGOT numbers, meaning their finishing was questionable, to say the least. Just look at what Emi Buendía and John McGinn did there:
One thing not to like is how well they defend set-pieces, but in the bigger picture, it doesn’t matter too much. By the way, the table of the best xGC performers in set-pieces so far is funny – Arsenal, then… Coventry, Hull, and Ipswich. Someone did their homework.
The bottom line is that allowing 5 xG over three games and conceding zero actual goals simply isn’t sustainable long-term. That’s the fact. Could that streak continue against Chelsea? It could, as much weirder things have happened in football.
But you’d need to have strong conviction to bet on it.
2️⃣ Is Cole Palmer even the best captain in GW4?
When we talk pure EV (expected value/expected points), it’s still just a guessing game, though an educated one.
He’s already taken 12 shots, and even though most weren’t particularly dangerous, that volume is a positive signal.
It’s always annoying to work with tiny sample sizes, but advanced metrics paint an optimistic picture – he posts elite numbers in goal probability added running metrics, meaning his carrying actually moves the ball into dangerous areas, and he has a great success rate 1v1.
The eye test certainly shows he’s moving better than last year, but it’s nice to see tangible numbers. That he can move pain-free and wherever he likes, with Xabi Alonso’s permission, means he can constantly receive the ball in dangerous positions. He’s dangerous both on and off the ball.
He also always plays 90 minutes, and you can expect him to take penalties, like the other night. Oh, and did I mention he faces Hull at home?
“So many routes to the points” depicted by FPL Copilot:
Even though we still don’t know whether we’re seeing the Palmer 23/24 or 24/25 version, we can reasonably expect that with João Pedro, Morgan Rogers, and Xabi Alonso along, the ceiling is there.
It’s a gameweek where the captaincy won’t be one-sided, because a lot of people will still captain Erling Haaland and Pedro, so there’s potential for a nice rank boost from Palmer without risking too much, especially when you already own that FWD duo.
There’s probably no reason (unless you have a good one) to play it on Rogers, because it’s the same “uncertain” situation as with Palmer, just with a lower ceiling and less security. Owning him is enough.
Pedro is fine. We know exactly how he performs, and he’s great again this season – the low sample size shouldn’t bother you in this specific situation.
3️⃣ Last week, I bought Cherki/Foden/Semenyo, and it failed miserably. Why should I buy Chelsea now?
Well, that’s an entirely different spot.
The City players you bought were much more punty than the Chelsea picks. There are many more xMin problems with players like Cherki or Foden, whereas Chelsea are more of a long-term project:
Cole Palmer: 3x 90 minutes
Morgan Rogers: 82, 90, and 87 minutes
João Pedro: 3x 90 minutes
So far, everyone is performing well, and I see no reason for that to change.
They also don’t play any European football, and when you look at their fixture run, there’s no sell-pressure any time soon.
Look at this Solio Analytics fixture ticker, ranked by predicted goal difference:
You can hold them for pretty much as long as you like.
4️⃣ Should I triple captain Palmer?
Stop overcomplicating that chip.
I’m fascinated by how much time people spend overanalysing when to use the Triple Captain and its outcomes, while neglecting other, MUCH more important stuff. Mass hysteria where it’s unnecessary.
You’ll triple-captain someone with 7 to 8.5 EV in the first half of the season in a single gameweek. So pick someone based on other factors, such as when you play other chips, your risk tolerance, and others.
If you want to be different and you’re not interested in triple-captaining Haaland (or you don’t want to own him in certain stretches of the season), you will probably use it on Bruno Fernandes, Bukayo Saka, or Cole Palmer. This is a chance to play it on someone with a nice balance of still relatively low effective ownership and high EV.
Simple as that.
5️⃣ Is Morgan Rogers a good pick?
I spent an unhealthy amount of time discussing Morgan Rogers during the pre-season, so just briefly.
For a complete Rogers analysis, I recommend you read the article above (free)
He won’t score bangers all the time, but he reminded us not long ago that he will, from time to time.
As with everything in football data analytics, you need context. And context is that he has 6 seasons of solid finishing under his belt, despite not playing in his best position most of the time, which limited his ability to take his best shots.
He’s now firmly installed in his most natural role.
He’s got better teammates.
And I’m still bullish on Xabi Alonso.
You can expect his chance-creation threat won’t go anywhere. Goals will follow, in some form.
For £7.6 million, he’s perfectly fine if he fits your plans. I don’t know, as I don’t see your team. But in total isolation, there’s nothing wrong with him.
That’s all for today.
See you very soon with Wildcard 4 draft, Free Hit 4 draft, and analysis of various players and teams!
