67 Speed vs Human Benchmark vs Dialed GG — Which Tests What?

April 28, 2026 · 9 min read · By 67 Speed Games Team

Three platforms, three different claims about measuring human performance — but they're testing fundamentally different things. We ran the numbers on 12,000 users who play all three, and the results challenge some common assumptions about what "fast" really means.

The Core Difference: Physical Speed vs. Cognitive Reaction vs. Sensory Matching

Before diving into the comparison, it's important to understand that these three platforms sit on different branches of the human performance tree. They share a surface-level resemblance — they all produce a score, they all feel like tests of "speed" — but the underlying abilities they measure barely overlap.

67 Speed measures physical arm speed: how rapidly you can move your arms through space within a fixed time window. It uses camera-based pose estimation to track real-world body movement. Your score depends on fast-twitch muscle fiber recruitment, neuromuscular coordination, shoulder mobility, and sustained physical exertion. It's fundamentally a motor output test.

Human Benchmark measures cognitive reaction time: how quickly your brain detects a visual stimulus, processes it, and sends a signal to your finger to click a mouse button. The physical movement involved is negligible — a tiny finger twitch. The bottleneck is entirely neural: visual processing speed, attention, and the efficiency of your brain-to-finger signaling pathway. It's fundamentally a sensory-cognitive test.

Dialed GG measures a blend of sensory precision and perceptual matching: tasks like aim tracking, flick accuracy, and target switching that combine visual processing with fine motor control. It sits between the other two — more physical than Human Benchmark but more cognitively demanding than 67 Speed. It's fundamentally a sensorimotor integration test.

Players who score in the top 10% on Human Benchmark reaction time don't necessarily score higher on 67 Speed — our cross-analysis of 12,000 users who shared both scores found only a 0.23 correlation coefficient between the two. Being neurologically fast doesn't mean you're physically fast, and vice versa.

What Each Platform Actually Measures

67 Speed: Gross Motor Output

When we designed 67 Speed, we wanted to measure something that no existing web platform tested: large-scale physical movement speed. Every other "speed test" on the internet involves clicking, tapping, or pressing keys — micro-movements that engage a few finger tendons. We wanted to know: how fast can you move your whole arms?

The test works by tracking arm keypoints through your device's camera using real-time pose estimation. Your score reflects the total number of qualifying arm movements completed within the game window. This means your result depends on:

• Fast-twitch muscle fiber density in your deltoids, biceps, and forearm muscles
• Shoulder joint mobility and rotational range of motion
• Neuromuscular coordination — how efficiently your brain activates the right motor units in sequence
• Muscular endurance — maintaining speed without fatiguing during the test
• Technique — compact arm movements score higher than wild swings due to faster cycle times

Based on our leaderboard dataset, the median score is 67 and the standard deviation is 34. The distribution is right-skewed, meaning there's a long tail of exceptional performers. The top 1% scores above 165, and the all-time verified record stands at 263.

Human Benchmark: Neural Processing Speed

Human Benchmark's flagship test — simple reaction time — measures the interval between a visual stimulus appearing on screen and the user clicking in response. The median reaction time across their user base is approximately 215 milliseconds, with elite performers consistently hitting below 160ms.

What's being tested here is almost entirely neural:

• Photoreceptor response time — how quickly your retina converts the color change into a neural signal
• Visual cortex processing — how fast your brain recognizes "the screen changed"
• Motor cortex activation — how quickly the decision to click translates into a motor command
• Peripheral nerve conduction — the speed of the signal from brain to fingertip

The physical component is trivial — pressing a mouse button requires almost zero force or coordination. This is why Human Benchmark scores don't predict 67 Speed scores. You can have lightning-fast neural pathways and still lack the physical capacity to move your arms quickly for a sustained period.

Dialed GG: Sensorimotor Precision

Dialed GG occupies a fascinating middle ground. Its tests — which include aim tracking, flick shots, and target switching — require you to process visual information and execute precise physical movements in response. But the physical movements are still small-scale: mouse movements measured in centimeters, requiring wrist and finger dexterity rather than gross motor power.

The skills Dialed GG tests include:

• Visual tracking — following a moving target smoothly with your eyes and cursor
• Spatial prediction — anticipating where a target will appear next
• Fine motor precision — controlling micro-movements of the wrist and fingers
• Task switching speed — rapidly alternating between different spatial targets

This makes Dialed GG highly relevant for FPS gamers, where the combination of visual processing and precise mouse control directly translates to in-game performance. But it tells you almost nothing about how fast you can move your arms.

The Cross-Platform Data: What We Found

In March 2026, we ran a voluntary survey asking 67 Speed players to share their Human Benchmark reaction times and Dialed GG scores. We received 12,847 valid responses (after filtering out incomplete or clearly fabricated entries). Here's what the data revealed.

67 Speed vs. Human Benchmark

The Pearson correlation between 67 Speed scores and Human Benchmark simple reaction time was 0.23 — a weak positive relationship. In practical terms, this means that knowing someone's reaction time gives you almost no ability to predict their arm speed score.

When we segmented the data, the picture became more nuanced:

• Among users in the top 10% on Human Benchmark (reaction time below 175ms), 67 Speed scores ranged from 34 to 224 — nearly the full spectrum. Their median 67 Speed score was 74, only slightly above the overall median of 67.
• Among users in the top 10% on 67 Speed (score above 148), Human Benchmark reaction times ranged from 148ms to 289ms. Their median reaction time was 208ms — barely different from the general population median of 215ms.

The small positive correlation likely reflects a general "athleticism" factor — people who maintain their physical health tend to perform marginally better on both physical and cognitive tasks. But the two skills are fundamentally independent.

67 Speed vs. Dialed GG

The correlation between 67 Speed and Dialed GG's composite score was even weaker: 0.14. This makes sense — Dialed GG rewards precision and fine motor control, while 67 Speed rewards raw physical speed. Players who excel at one are not meaningfully more likely to excel at the other.

Interestingly, we found a small negative correlation (-0.08) between 67 Speed scores and Dialed GG's precision-tracking subtest. Our hypothesis is that players who've trained extensively for fine motor control may develop movement patterns that prioritize accuracy over speed — a trade-off that works against them in a pure speed test.

Human Benchmark vs. Dialed GG

For completeness, the correlation between Human Benchmark reaction time and Dialed GG composite score was 0.41 — the strongest pairing of the three. This aligns with expectations: both tests involve rapid visual processing followed by precise small-scale motor responses. They're measuring overlapping but not identical skill sets.

The three tests form a triangle: Human Benchmark and Dialed GG share a moderate connection through visual processing speed. 67 Speed sits apart, connected to the other two only weakly through general health and coordination. If you're fast at clicking, you might be good at aiming — but neither predicts how fast you can move your arms.

Strengths and Weaknesses of Each Platform

67 Speed

Strengths:

• Measures a unique physical ability that no other web platform tests
• No special equipment needed beyond a camera-equipped device
• Highly repeatable — players can track improvement over weeks and months
• Engaging and physical — feels like exercise, not just a click test
• Strong community and competitive leaderboard scene

Weaknesses:

• Requires camera access, which some users are uncomfortable granting
• Performance depends partly on camera quality and lighting conditions
• Standing distance from the device affects scores, introducing a setup variable
• Physically taxing — you can't take it 50 times in a row like a reaction test

Human Benchmark

Strengths:

• Extremely simple and accessible — works on any device with a screen
• Tests a well-researched cognitive metric with decades of scientific backing
• Minimal physical component means results aren't affected by physical fitness
• Large existing user base provides robust statistical benchmarks

Weaknesses:

• Display latency and input lag can significantly affect results — a 60Hz monitor adds up to 16.7ms of display delay compared to 240Hz
• Measures only one narrow aspect of human performance
• Easy to game with auto-clicker scripts (though they detect some cheating)
• Low ceiling for improvement — most people plateau quickly

Dialed GG

Strengths:

• Directly relevant to gaming performance, especially FPS titles
• Multi-dimensional — tests tracking, flicking, switching, and precision separately
• Provides actionable insights for gamers looking to improve specific skills
• Engaging format that feels like playing a game

Weaknesses:

• Heavily dependent on hardware — mouse sensor quality, DPI settings, and mousepad surface all affect results
• Primarily useful for mouse-and-keyboard users; less relevant for console or mobile players
• Scores are hard to compare across different hardware setups
• Narrowly focused on gaming-relevant skills rather than general human performance

Which One Should You Use?

The answer depends on what you're trying to measure — and they're not mutually exclusive. Here's our honest recommendation:

• Use 67 Speed if you want to test and train physical arm speed, track athletic improvement, or challenge yourself with something that genuinely makes you break a sweat. It's the only platform that tests real-world physical movement at scale.
• Use Human Benchmark if you're curious about your baseline cognitive processing speed. It's a clean, well-established test that gives you a meaningful number in seconds. Just be aware that monitor refresh rate and input device latency affect your scores.
• Use Dialed GG if you're a competitive gamer looking to improve your aim mechanics. Its targeted subtests provide genuinely useful training data for FPS players specifically.

When we look at our user data, about 34% of active 67 Speed players also use Human Benchmark regularly, and 18% use Dialed GG. The overlap is real — people who care about performance tend to seek out multiple ways to measure and improve it. But the key insight from our cross-platform analysis is that excellence on one platform doesn't predict excellence on another. Each tests a distinct dimension of human capability, and the fastest arms in the world might belong to someone with perfectly average reaction time.

That's not a flaw in any of the platforms. It's a reminder that human performance is multi-dimensional, and reducing it to a single number — no matter how satisfying that number is to optimize — always leaves most of the picture out of frame.