Citation [citation='GeorgeFloydOverdosed' commentaire='53680'] 86 % | Eduardo Camavinga (DM) = 3e, 5e, 2e, 3e, 4e = 3,4 position 73 % | Andrian Kraev (DM) = 9e, 8e, 5e, 5e = 6,75 position
Rappelons que 87 % pour ST équivaut à 2,571 % et 73 % pour ST est à 6,111, et que les positions autres que ST semblent constamment ~1 position plus basses.
Chaque position pour la formation Knap 424 a été échantillonnée maintenant, et je suis satisfait que le même ensemble d’attributs requis s’applique à toutes les positions de champ extérieur de la même manière.
J’ai fait quelques ajustements des pondérations.
J’ai laissé tomber Sterling l’équivalent de ~1 position, mais cela signifie qu’il devrait s’attendre à une position de ~4, pourtant sa moyenne est de 5,375 si on donne +1 pour ne pas être ST. C’est le mieux que j’ai pu faire.
Les positions non-ST semblent désormais plus comparables aux ST (moins de différence de position inexpliquée), et l’endurance est désormais plus raisonnable à 40 au lieu de 94 - l’endurance semblait être la cause partielle de la surévaluation de Sterling. Beaucoup de joueurs ont des changements agréables et fortuits, comme Maeda qui monte plus haut et Gyokeres redescend, ce qui correspond davantage au classement réel.
Cependant, certains changements ont dû aller avec, comme la chute de Vlahovic un peu en dessous de Messi. Je pense qu’il est plus probable que Sterling perde 1 position, que Vlahovic ne perd pas 0,5 position. Et comme la plupart des joueurs semblent plus précis aujourd’hui, je ne m’inquiète pas trop pour Vlahovic.
L’étape suivante consiste à traduire en poids Genie Scout, puis je pourrai télécharger le fichier mis à jour. La principale raison de la mise à jour du fichier sera bien sûr d’ajouter les nouveaux poids de gardien. [/citation]
Reading back through this thread's evolution, a few points seem worth flagging clearly — not to dismiss the overall contribution (the physical-attributes-dominate thesis holds up well against Zippo's isolated A/B test and skawkclsrn's large-scale randomized study), but because several specific claims here don't hold up as well as they're presented:
The "position proficiency caps at 18" claim (repeated with confidence early on) turned out to be wrong once tested — worth flagging since it may still be floating around in older posts/quotes. The Genie Scout personality-weighting bug wasn't caught internally — it took LightningFlik's independent tool to demonstrate the Osimhen-over-Haaland inconsistency. Worth being more explicit that GS output shouldn't be trusted at face value for hidden attributes even now. The "Blended" file and the "15 pace/acc realistic template" contradict each other on your own numbers (the Newcastle-vs-synthetic-squad comparison) — this was acknowledged but never really resolved, just reframed as "the file is a rough guesstimate." That's a fair characterization, but it should probably be stated up front rather than discovered by a reader cross-checking the math. The "attribute thresholds are relative to competition level" theory, used to explain away discrepancies with ykykyk05251/Zippo/Orion/HarvestGreen22, has never actually been tested — and skawkclsrn's much larger randomized dataset found no thresholds anywhere, which cuts against it rather than for it. A lot of the decimal-precision weights (Pressure=52, Determination=56, etc.) come from single small-sample runs, some explicitly labeled "pure speculation" or "I don't remember why I set this" in your own posts. That's fine as a starting point, but the precision of the numbers implies a confidence level the underlying data doesn't really support.
None of this undercuts the value of the overall synthesis — it's clearly the most complete one on the forum. But the granular numbers probably deserve a "confidence level" caveat next to them, given how often they've had to be walked back or reframed.
A4 said: I read this thread and it got me wondering: are reflexes and agility basically everything you need to make a good GK? Also, what's the best way to achieve a high match rating? Is there any way to influence it other than saving lots of shots and keeping clean sheets?
What's the best role or team instructions for farming GK ratings? How much do traits affect a goalkeeper's performance? Is height really that important?
Is there a way to build tactics around the goalkeeper being more involved in play and actually doing things? For example, something similar to the long-throw abuse in the old El Diablo tactics.
You can shut your mouth please When you're there doing nothing, don't criticize the work of other people, you obsessed We need people like george and haverstgrenn who move, who do a lot of testing and who take a lot of their time to write and explain to us the interpretation of their research, not obsessed people like you, you obsessed, leave us in peace, the obsessed.
86 % | Eduardo Camavinga (DM) = 3e, 5e, 2e, 3e, 4e = 3,4 position
73 % | Andrian Kraev (DM) = 9e, 8e, 5e, 5e = 6,75 position
Rappelons que 87 % pour ST équivaut à 2,571 % et 73 % pour ST est à 6,111, et que les positions autres que ST semblent constamment ~1 position plus basses.
Chaque position pour la formation Knap 424 a été échantillonnée maintenant, et je suis satisfait que le même ensemble d’attributs requis s’applique à toutes les positions de champ extérieur de la même manière.
J’ai fait quelques ajustements des pondérations.
J’ai laissé tomber Sterling l’équivalent de ~1 position, mais cela signifie qu’il devrait s’attendre à une position de ~4, pourtant sa moyenne est de 5,375 si on donne +1 pour ne pas être ST. C’est le mieux que j’ai pu faire.
Les positions non-ST semblent désormais plus comparables aux ST (moins de différence de position inexpliquée), et l’endurance est désormais plus raisonnable à 40 au lieu de 94 - l’endurance semblait être la cause partielle de la surévaluation de Sterling. Beaucoup de joueurs ont des changements agréables et fortuits, comme Maeda qui monte plus haut et Gyokeres redescend, ce qui correspond davantage au classement réel.
Cependant, certains changements ont dû aller avec, comme la chute de Vlahovic un peu en dessous de Messi. Je pense qu’il est plus probable que Sterling perde 1 position, que Vlahovic ne perd pas 0,5 position. Et comme la plupart des joueurs semblent plus précis aujourd’hui, je ne m’inquiète pas trop pour Vlahovic.
L’étape suivante consiste à traduire en poids Genie Scout, puis je pourrai télécharger le fichier mis à jour. La principale raison de la mise à jour du fichier sera bien sûr d’ajouter les nouveaux poids de gardien.
[/citation]
Reading back through this thread's evolution, a few points seem worth flagging clearly — not to dismiss the overall contribution (the physical-attributes-dominate thesis holds up well against Zippo's isolated A/B test and skawkclsrn's large-scale randomized study), but because several specific claims here don't hold up as well as they're presented:
The "position proficiency caps at 18" claim (repeated with confidence early on) turned out to be wrong once tested — worth flagging since it may still be floating around in older posts/quotes.
The Genie Scout personality-weighting bug wasn't caught internally — it took LightningFlik's independent tool to demonstrate the Osimhen-over-Haaland inconsistency. Worth being more explicit that GS output shouldn't be trusted at face value for hidden attributes even now.
The "Blended" file and the "15 pace/acc realistic template" contradict each other on your own numbers (the Newcastle-vs-synthetic-squad comparison) — this was acknowledged but never really resolved, just reframed as "the file is a rough guesstimate." That's a fair characterization, but it should probably be stated up front rather than discovered by a reader cross-checking the math.
The "attribute thresholds are relative to competition level" theory, used to explain away discrepancies with ykykyk05251/Zippo/Orion/HarvestGreen22, has never actually been tested — and skawkclsrn's much larger randomized dataset found no thresholds anywhere, which cuts against it rather than for it.
A lot of the decimal-precision weights (Pressure=52, Determination=56, etc.) come from single small-sample runs, some explicitly labeled "pure speculation" or "I don't remember why I set this" in your own posts. That's fine as a starting point, but the precision of the numbers implies a confidence level the underlying data doesn't really support.
None of this undercuts the value of the overall synthesis — it's clearly the most complete one on the forum. But the granular numbers probably deserve a "confidence level" caveat next to them, given how often they've had to be walked back or reframed.
What's the best role or team instructions for farming GK ratings? How much do traits affect a goalkeeper's performance? Is height really that important?
Is there a way to build tactics around the goalkeeper being more involved in play and actually doing things? For example, something similar to the long-throw abuse in the old El Diablo tactics.
https://fm-arena.com/thread/20264-more-attribute-weight-findings-through-data-mining-fm26/page-1/
reflexe and agility
When you're there doing nothing, don't criticize the work of other people, you obsessed
We need people like george and haverstgrenn who move, who do a lot of testing and who take a lot of their time to write and explain to us the interpretation of their research, not obsessed people like you, you obsessed, leave us in peace, the obsessed.