The 25-point cognitive gap between Gold and Diamond isn't aim. It's decision speed.

An early-signal look at 140 NeuroRank combines. The dimensions that separate Gold-band and Diamond-band profiles are not the ones most players train.

---

Data Thursday is back. This week we sliced our combine cohort by overall cognitive score and ran the same question every player asks us: what actually changes between a Gold-tier brain and a Diamond-tier brain?

The answer is not aim. Aim is the dimension that looks most saturated across tiers. The two dimensions that open up a real gap are decision quality and raw reaction speed, and the effect size is large enough that we are flagging it now, with caveats, because the pattern is consistent across every way we slice it.

Here is the headline number. On decision quality, the median Gold-band player scores 60. The median Diamond-band player scores 95. That is a 35-point median gap on a percentile scale, in a 100-point range. It is the largest separator of any core dimension we measure.

Let's get into it.

---

How we defined the tiers

NeuroRank does not ask players for their in-game rank. We measure cognitive performance directly, then assign an overall percentile score and archetype. For this post we mapped cohort overall scores onto familiar ladder names so the comparison reads cleanly:

Band	Overall score range	N in cohort
Bronze	below 40	2
Silver	40 to 49	8
Gold	50 to 64	54
Platinum	65 to 74	61
Diamond	75 and above	15

Gold is the largest band. Diamond is the sparsest, which is why everything below carries an early-signal caveat: the Diamond sample is 15 profiles, large enough to see directional truth, too small to nail down third-decimal effect sizes. We will re-run this post at 500 and 5,000 combines.

---

1. Decision quality is where tiers actually separate

Every NeuroRank combine measures decision quality through the Composure (Flanker) and Tilt modules. It captures how a player thinks when the screen is noisy and the stakes are rising.

On this one dimension, the median Gold-band player lands at the 60th percentile. The median Diamond-band player lands at the 95th. 80% of our Diamond cohort scored above the 75th percentile on decision quality. Only 38.9% of Gold players cleared the same bar.

The mean-gap story is the same. Gold averages 61.4. Diamond averages 86.6. That is a 25.2-point mean difference, Cohen's d of 1.08. In behavioural data, a d above 0.8 is considered a large effect. Decision quality clears it comfortably.

---

2. Raw reaction speed is the second separator, and it's the opposite of what players assume

Here is a result that surprises most players we have shown it to: Gold-band combines have a median raw speed score of 37.5. Diamond-band combines have a median of 60. Diamond players are not wildly fast. They are reliably above average. Only 26.7% of Diamond profiles scored above the 75th percentile on raw speed. That number drops to 7.4% in Gold.

Raw reaction is the easiest dimension to grind and the hardest to cap out. What the data shows is that you do not need to be elite to break into Diamond. You need to stop being bad. Gold-band players leak the most here: half of them score in the bottom 40 on raw speed. That is the single most consistent weak spot we see in the middle of the ladder.

---

3. Aim is almost fully saturated by Gold

Now the deflating number for anyone whose training routine is 90% aim trainer. Median aim precision in Gold is 71. Median aim precision in Diamond is 79. The mean gap is 7.6 points, the smallest of any core dimension.

72.7% of Diamond players score above the 75th percentile on aim. 22.6% of Gold players already clear the same bar. That means roughly one in four Gold players already aims at a Diamond level. What is keeping them in Gold is not their mouse hand.

Saturation on aim is consistent with what our archetype data has shown for months. Once a player crosses the 70th percentile on aim precision, further gains on that one dimension stop moving their overall score. Aim is the clearest case of the ceiling effect we will cover later in this rotation.

---

4. The full Gold-to-Diamond table

Dimension	Gold median	Diamond median	Mean gap	Cohen's d
Decision quality	60	95	+25.2	1.08
Raw speed	37.5	60	+22.6	1.21
Tracking accuracy	65.5	76	+10.8	0.89
Consistency	60	75	+9.1	0.75
Aim precision	71	79	+7.6	1.25

Aim precision has a small absolute gap but the highest standardized effect size, because the variance in Gold is tiny. Diamond players aim a bit better than Gold players with less spread. The real-world takeaway stays the same: aim is not where the ladder opens up.

---

5. Breadth, not spike

Here is the structural difference we did not expect to find this cleanly.

Gold-band profiles usually have a spike. 79.6% of Gold players scored above the 75th percentile on at least one dimension. They are not cognitively weak. They are cognitively lopsided. A Gold profile typically has one strong dimension carrying a weak one.

Diamond-band profiles look different. 18.2% of our Diamond cohort scored above the 65th percentile on all five core dimensions. Only 1.9% of Gold did. That is a roughly 9.5x ratio, on a full-floor threshold, in a small sample. Diamond players do not necessarily have one elite spike. They have no holes.

If you want a mental model: Gold looks like a staircase with one missing step. Diamond looks like a short, flat platform.

---

6. What this says about training

Three patterns the data keeps repeating:

1. Training decision quality has the highest slope. It moves the most between tiers, and most players ignore it because it is harder to feel than aim.
2. Raw reaction speed is a remedial fix, not an elite one. Once you clear the 50th percentile, training it further yields less than improving your weakest dimension.
3. Fix the hole, not the spike. If you are a Gold player with one dimension above the 75th percentile and one below the 40th, your overall score will move faster from raising the floor than pushing the ceiling.

These conclusions match what our archetype classifier has been doing under the hood all along. The Balanced Operator archetype, which makes up 23.6% of our cohort, shows the highest median overall score. Specialists who spike on a single dimension rarely climb past Platinum, even with elite aim or elite reaction.

---

Methodology

We analyzed 163 total profiles from the NeuroRank combine, of which 140 had a complete overall cognitive score as of April 17, 2026. All statistics above are aggregates over groups of at least 11 profiles. No individual player data appears in this post. Tier bands are assigned by NeuroRank's own overall cognitive percentile, not by self-reported in-game rank. Cohen's d is pooled across both samples. Because our Diamond-band sample is 15 profiles, we are calling this early signal rather than definitive. We will re-run the analysis at 500 and 5,000 completed combines and publish any reversals. Scoring definitions for each dimension are frozen and were not changed for this analysis.

---

Take the combine

If you want to know which side of this gap you are on, run the six-module NeuroRank combine. It takes about 25 minutes, measures all five dimensions above plus two more, and returns an archetype and a written scouting report.

Run the combine

If you have already taken it, re-run it in 30 days and watch which dimension moves. The players who climb fastest on our retest data are the ones who raise their floor, not the ones who chase a higher spike.