What is transpiring at Hearts this season is an excellent case study on performance measurement and related attribution now that they are 26 games into the league season, which was the tenure of Neil Critchley last season.
Firstly, Tony Bloom said he expected Hearts to compete
Over the past 7 seasons prior to 2025-2026, the 2nd place team had an average non-penalty xG difference of 1.07 and goal difference of 1.36. The lowest 2nd place finisher in each were both Rangers last season, with an non-penalty xG difference
about double that of Hearts so far this season, with a goal difference per game slightly below Hearts'.
That gap between goal difference and underlying performance levels is where randomness lives, which is often misattributed to managers.
Hearts' goal difference through
26 games of +28 is 23 more than the +5 during Critchley's 26 games last season.
A pretty simple attribution framework results in the following in order of impact:
Own goals and penalties +10
Finishing +8
non-penalty xG +3
Keeper shot stopping +2
Those number represent the net change and impact between for and against, so this season's Hearts team has benefited by 10 goals relative to the 26 games under Critchley between own goals and penalties for vs against, for example.
Clubs like Brentford and those affiliated with Bloom very likely understand these dynamics, and also recognize that "dumb money" misattributes a lot of the swings to managers and players.
They'll assess managers similarly to the now infamous Nancy Venn Diagram.
If Bloom had/has an expectation of this season's Hearts' squad being good enough to "at least" finish 2nd, then that suggests an underlying performance gap of at least 0.4 non-penalty xG so far.
Of the roughly 3 goals in improved underlying performance level, despite the massive
overhaul in the playing squad via Jamestown Analytics, 2 are from set pieces and just 1 from open play.
Given Hearts hired a dedicated set piece coach, this leaves a 1 goal increase that is more directly attributable to the head coach.
To match the worst 2nd place team from
open play, Hearts would have to have increased over the level under Critchley by 13 to 14 expected goal difference, which compares to 1 through 26 games.
That -12 to -13 is a decent performance scorecard for Derrick McInnes, and even if Hearts win the league, which is
clearly plausible, the things he can reasonably control are likely performing WAAAY below Bloom's expectations.
The vast majority of people disagreeing with this kind of analysis is precisely how Bloom and Behnam made hundreds of millions betting against the majority.
@threadreaderapp unroll
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Edinburgh Derby Saturday presents an opportunity to flush out a bit on Wilfried Nancy's recent comments on "luck."
The concept of variance is one to which I have returned regularly the past few seasons, including trying to break out variability between factors teams control vs
those they do not.
For example, teams control, at least indirectly, the level of player they employ as strikers and keepers, which impact finishing and shot stopping. There is variability around those factors, but there is a degree of agency.
Examples of uncontrollable factors
are the randomness of how opposition players finish and shot stop against a team across time.
The image above was one I posted in June 2024 showing how much Hearts had benefited from uncontrollable factors during 2023-2024 season, which often gets misattributed to managers
Mostly keep this kind of info hidden away from delicate sensibilities over at @huddlebreakdown these days but desperate times and all that.
Sorry - it's a lot of games so you'll have to minimize/maximize
The Cup Final was bad, and was critical of various elements of selection, tactics, substitutions. Roma was not much different than Atalanta last season...only Atalanta couldn't hit the broad side of a barn that day, so goats became heroes.
The disparity in goals vs
xG includes the Hearts and Dundee Utd games also having 1 goal in each from Schrodinger's offside set pieces but none for Scales OG.
The two worse 2-game avg in the graphic were Oct 1 2021 near the Ange depths and the other during the horrible stretch in December 2023 under
This was shortly after an article doing a deep dive on Rangers' analytics setup at the time was published in The Athletic and I had stopped doing Good Bad & Ugly game threads.
Coming up on 4 years later and Thelwell's just-released interview makes it likely they are still
As referenced, the hiring of Jaymes Monte an indication they are looking to progress, and I guess it is plausible the "manager has final say on recruitment" framing could be a PR strategy.
Stopped GBU once I was convinced Ange had the team
set to perform at a high level and "Bad & Ugly" parts of threads would have to recycle long-term structural issues that persist to today.
Being optimistic on a potential Nancy hire does not mean de facto expectation that structural issues are being addressed.
I am not suggesting something similar is likely or imminent, but it is a heuristic for what I keep chronicling relative to McInnes and Hearts.
Huge difference is Hearts' executive team are new to the operating model and probably
complicit in deviating from it, as was evidenced at the commencement with how they went about replacing Critchley.
Clubs like Midtjylland trust the operating model, with good reason, with supremacy. The components, whether DoF or head coach, have to be aligned or else.
Sentiment is understandably skeptical on every move, but cleaning house of legacy scouting and recruitment personnel is a good sign on a relative basis, IMO.
Even with the DD-as-de-facto-DoF the quality of talent ID and profiling has been erratic and poor.
It is possible that DD has finally woken up a bit on a relative basis?
In a column from July where I reviewed his September 2020 interview published in The Athletic, he referenced his view on collective responsibility and reticence to fire people.
Is he going to fire himself?
With many in the support talking about the club "modernizing," what would that look like?
Directionally, it would be hiring someone like Wilfried Nancy, who currently works in one of the more progressive and emerging analytics-focused clubs in MLS, Columbus.
Our @huddlebreakdown audience can attest to the fact that I began commenting that Hearts could have a legit opportunity to win the 2025-2026 season back in February.
I even bet some money on them when they were at 25-1 odds...obviously not because I am "rooting" for them
but because I thought it was badly mispriced...well McKinlay's interview with @mcgowan_stephen makes me feel better about recent decision to have already banked that mispricing.
He speaks with a forked toungue, as has McInnes at times in public. They seem to be engaged in some
combination of preference falsification and/or self-delusion amidst an appeal to the Bloom "authority".
@Alan_Morrison67 and I laid out two anecdotal sign posts on how progressive the adoption and integration of the Bloom revolution would be within the culture at Hearts.