Editorial Intelligence vs FPL, part five: Waiting has a cost too

Part of Editorial Intelligence vs FPL, a running experiment in evidence, judgement, football and occasionally making terrible decisions.

I have just taken a four-point hit in Fantasy Football to bring in Alexander Isak.

That is already slightly uncomfortable, because the entire point of this experiment has been to resist the temptation to turn the latest result into an instruction.

Isak has started the season strongly. Liverpool have Fulham at home next. He looks like an obvious target. But the interesting part of the decision wasn’t whether I liked him. It was when to act.

The structure I wanted was simple: Bruno Fernandes + Igor Jesus out, Cole Palmer + Isak in. It fitted my budget, leaving £0.5m in the bank.

Just.

If Isak rises in price, that structure becomes more expensive and potentially disappears. Waiting gives me more information about injuries, team news and the next deadline. Acting early protects the option, but exposes me to almost a week of new uncertainty.

So I made the move.

Waiting is not neutral

AI systems are very good at recommending more evidence.

Check the team news. Wait for press conferences. Get another data point. Reassess closer to the deadline. Reduce uncertainty before you decide.

Usually that sounds sensible because it is sensible. But there is a hidden assumption inside it: that waiting is free. It isn’t always.

In this case, waiting created its own risk. The player could become more expensive. My preferred transfer structure could become less affordable. The option I had today might not exist later in the week.

So the choice wasn’t between uncertain action and safe inaction.

It was between two different kinds of uncertainty:

• act early and accept injury, rotation and team-news risk;
• wait and accept price and opportunity-cost risk.

That is a much more useful decision problem.

The four-point hit is part of the decision

It would be easy to tell the Isak story as a reaction to form. He has scored goals, therefore buy him.

That isn’t quite what happened.

The move costs me four Fantasy points before Isak or Palmer have kicked a ball for me. That means they do not merely need to look attractive in isolation. The new structure needs to justify paying to get there now rather than waiting for another free transfer.

That is why I want the rationale written down before Gameweek 4 starts.

The decision is:

Bruno Fernandes → Cole Palmer

Igor Jesus → Alexander Isak

Cost: -4 points

Bank remaining: £0.5m

The football case is that Palmer has Hull at home, Isak has Fulham at Anfield, and the move shifts money away from an expensive Manchester United asset before a much tougher fixture while upgrading a forward position that had produced very little.

The execution case is different: I acted early because the structure was affordable now and I did not want a price rise to remove that flexibility.

Those are related decisions, but they are not the same decision.

This is not a bet that Isak will score

That distinction matters because next weekend will make hindsight very tempting. If Isak scores twice, this will look obvious. If he blanks, gets injured in training or Palmer does nothing against Hull, the four-point hit will look reckless.

Neither result will tell me whether the decision was sensible at the point I made it. The useful audit is whether the evidence and trade-offs justified accepting one kind of risk to avoid another.

That is where this experiment keeps drifting away from Fantasy Football and back towards the work I am actually interested in.

In editorial and business decisions, we often behave as though more information is always better. More research. More stakeholder input. Another model. Another meeting. Another week to think.

Sometimes it is.

Sometimes delay changes the environment you are trying to make the decision inside.

The opportunity disappears. The cost increases. Somebody else moves first. The news cycle changes. The window closes.

At that point, “wait for more evidence” is no longer caution. It is a decision with consequences of its own.

The new question

Gameweek 2 taught me that AI could work as a decision brake. After a bad result, it helped stop me overreacting.

Gameweek 3 added the opposite lesson: if the system only ever tells me to wait, it isn’t judgement. It’s inertia.

Gameweek 4 makes that more concrete.

The question is no longer simply:

Do I have enough information to make the decision?

It is:

How much uncertainty am I willing to accept in order to preserve an opportunity?

That feels much closer to the decisions people actually make.

There is rarely a perfect answer sitting behind one more piece of research. There is usually a point where the value of additional information has to be weighed against the cost of waiting for it.

So I have made the gamble.

Palmer and Isak are in. Bruno Fernandes and Igor Jesus are out. Four points have already gone.

By next weekend I will know whether the move worked.

The more useful question will be whether it was a sensible gamble before I knew the result.

Editorial Intelligence 1. Fantasy Football 1.