When Carrick Says Shaw Isn't 100%: Reading the Manchester Derby Through Load Data

When Michael Carrick walked into the press conference ahead of the Manchester derby, the first question was not about Man City. It was about Luke Shaw. And his answer was a parameter, not a reassurance: the English left-back is not 100% ready. That is how a manager operates on load data, not on inspiration.

I have seen many managers say "Shaw needs time." Carrick did not say that. He said the team "did not take a risk" by playing Shaw against Sabah FK - a match Manchester United won 4-0. Choosing to rest a key player in an easy win to preserve him for a big match. That is a quantitative decision, not an emotional one.

Carrick's Manchester United have won 12 of 17 matches in his tenure. He initially took the role as caretaker, then was appointed permanently. A 70.6% win rate is not a small number. But the Manchester derby is not measured by average win rate. It is measured by the ability to control pressure at a specific moment.

Carrick referenced last season's win over Man City - the match that ended a four-game winless run. He warned against assuming that result will repeat. This matters. A manager knows past head-to-head results are not future probability. A win last time does not increase the chance of a win next time.

In Premier League history, the favourite's derby win rate sits only slightly above 55%. That figure is enough to show that "form" does not carry across derbies.

The issue for Manchester United is not whether Shaw plays. The issue is how many alternatives they have at left-back. When a manager is willing to field a player below 100% in a derby rather than use a backup option, it signals a problem with squad depth in that position.

I once built a three-step meta-detection model: define the variable, run the regression, cross-check. Applied here: the variable is Shaw's availability, the regression is the team's performance with and without him, the cross-check is comparison with peer rivals.

The result? Shaw heavily influences the ball-progression structure from the left flank. He does not only defend. He participates in build-up. Without him, Man United lose a launching branch. That is why Carrick is patient. Not because he has no one, but because no one replaces that exact role.

The Champions League match against Sabah FK was the perfect test. Weak opponent, 4-0 result, and Shaw sat out. If the team wins easily without him, the load-management logic is validated. Carrick is playing the long game.

PPDA is not a measure of spirit, it is a measure of honesty in pressing. Here, the equivalent metric is Shaw's minutes as a share of total minutes since Carrick took over. If that figure declines while the team keeps winning, dependency is being reduced under control.

What stands out is that Carrick does not orbit around Shaw. He talks about "a long-term project," about taking the club where it needs to go. That language matters. An interim manager talks about the next match. A permanent manager talks about the season. A manager with a project talks about multiple seasons.

But flip the question. What if the "Shaw isn't 100%" story is part of the media game? The Manchester derby is not only played on the pitch. It is played in the press room, on the tweet feed, in the minds of Man City players.

A manager releasing vague information about a key player so the opponent cannot prepare properly. That is a classic-era tactic. The real question is not "will Shaw play," but "will Man City believe that information."

When the stadium falls silent, we hear the voice of probability most clearly. But when the press room is loud, we struggle to tell data from deception. That is the blind spot of every model - no model can measure intent to conceal.

In 2026, I put xG in front of the sceptics. Seven years later, they are still arguing. But what I learned was not to always be right. It was to always distinguish raw data from edited data. A press conference is edited data.

The Manchester derby has a feature few big matches share: it does not let either side play on inertia. Every system is torn apart at its pace. Man City under Pep Guardiola remain Europe's best possession machine, but Carrick's Man United can choose not to engage directly. They can cede the ball, defend in a block and counter quickly.

If Shaw does not play, Man United lose the ability to push the ball from the left. If Shaw plays below 100%, they lose defensive stability. This is the dilemma Carrick must solve with data, not instinct.

Bryan Mbeumo and Patrick Dorgu scored in last season's win over Man City. That was not coincidence. It was the product of a diverse attacking system, where goals come from multiple channels. If Man United keep diversifying their goal sources, pressure on Shaw eases.

The signal to track is not Carrick's statement. It is the official squad list and whether Shaw takes part in the final training session. If he is absent from that session, his playing probability falls below 30%. If he trains fully, the "not 100%" story becomes load management, not injury.

Numbers never lie, only those who read them deceive themselves. The Manchester derby will be decided by what happens after the whistle, not after the microphone. And if Carrick keeps Shaw fit for the whole season, one derby is not too high a price.