The CS2 Sensitivity Formula 80% of Pros Use — and Why Your eDPI Is Probably Wrong

Verified against active professional play data from ProSettings.net, April 2026. Values may change with future CS2 patches.

The most common aim problem in CS2 isn’t bad mechanics — it’s wrong settings. Too much sensitivity and you overshoot every flick. Too little and you can’t react fast enough in close-range fights. Most guides paste a pro settings table and leave you to figure out the rest. This one explains the logic behind those numbers, which ones apply to your playstyle, and how to calibrate your own sensitivity so muscle memory can actually form.

Quick Start: Six Changes to Make Right Now

If you have five minutes before your next session, make these changes first — they eliminate the most common aim-killers before you touch anything advanced:

1. Disable Enhance Pointer Precision — Windows Control Panel → Mouse → Pointer Options → uncheck the box
2. Set Windows pointer speed to position 6 of 11 — the exact middle notch in the same panel
3. Enable Raw Input in CS2 — Settings → Game → Mouse → Raw Input: On (or open console and type: m_rawinput 1)
4. Disable Mouse Acceleration — CS2 console: m_customaccel 0, then host_writeconfig
5. Set polling rate to 1000Hz — in your mouse software (Logitech G Hub, Razer Synapse, ZOWIE utility)
6. Calculate your current eDPI — DPI × in-game sensitivity. If the result is under 600 or over 1400, you have calibration work to do

eDPI: The One Number That Actually Describes Your Aim

eDPI (effective DPI) is your mouse DPI multiplied by your CS2 in-game sensitivity. It describes how far your crosshair moves per inch of physical mouse movement — one unified number regardless of which DPI and sensitivity combination you use to reach it.

Why this matters: three settings that look completely different on paper are functionally identical:

• 400 DPI × 2.0 sensitivity = 800 eDPI
• 800 DPI × 1.0 sensitivity = 800 eDPI
• 1600 DPI × 0.5 sensitivity = 800 eDPI

The crosshair travels the same distance per inch in all three cases. The reason to prefer 400 or 800 DPI over higher values is hardware-level: most gaming mouse sensors are marginally more accurate at lower DPI values, though the gap matters far less than it did in the CS:GO era.

The professional eDPI cluster sits between 700 and 1200. Below 600, completing a 180-degree reaction turn requires more physical wrist movement than is practical under pressure. Above 1400, micro-adjustment precision drops enough to affect headshot consistency on distant targets — the physics works against you before your mechanics do.

DPI: Why Almost Everyone Uses 400 or 800

You have two reliable choices: 400 DPI or 800 DPI.

400 DPI is the historical default, dating to CS 1.6 when sensors at higher DPIs introduced measurable tracking jitter. Modern sensors (PixArt 3395, Logitech Hero 25K, SteelSeries TrueMove Air) are accurate at all DPIs above 400, but pros don’t change settings that win matches. Roughly 70% of tracked CS2 professionals use 400 DPI.

800 DPI is the practical alternative if your mousepad is under 30cm wide or if you want a higher base sensitivity without the in-game sensitivity number feeling unnatural. NiKo (720 eDPI) and donk (1000 eDPI) both use 800 DPI at the highest competitive level.

Avoid DPIs above 1600 for competitive play. At high values, many sensors use interpolation to synthesize precision the hardware doesn’t actually have. The result is micro-jitter during slow tracking that’s imperceptible on-screen but registers as inconsistency in aim over time.

Pro Player eDPI Reference Table

Data from ProSettings.net and Specs.gg, April 2026.

Four of six fall between 700 and 1000 eDPI. s1mple at 1236 is the high outlier — his wrist-dominant, high-aggression playstyle is the reason, not a random preference. All six use 1280×960 stretched.

Resolution: 1280×960 Stretched vs 1920×1080 Native

Five of six pros above use 1280×960 4:3 stretched. The mechanical reason: stretching a 4:3 image to fill a 16:9 panel widens everything horizontally. A player model measuring 60 pixels wide at 1920×1080 becomes roughly 80 pixels wide at 1280×960 stretched — a 33% larger target for click-aim. The FPS bonus is real but secondary: rendering at lower resolution produces higher average frame rates and lower input latency.

The cost is a narrower horizontal field of view. You see less of the map on the peripheral edges. For AWPers and aggressive entry fraggers — whose engagements are already directional — this is a favourable trade. For IGLs and support players who win rounds through map awareness, it isn’t.

Use 1280×960 stretched if: you are primarily an AWPer, rifler, or entry fragger, and you are above Gold Nova where deliberate crosshair placement has replaced reactive spraying.

Stay on 1920×1080 native if: you are Silver through Gold Nova (wider FOV teaches positioning faster than bigger hitboxes), or you are an IGL or lurker who depends on peripheral information.

To enable stretched res: CS2 Settings → Video → Resolution: 1280×960 → Aspect Ratio: 4:3 → Display Mode: Fullscreen. If black bars appear, set your GPU scaling mode to Fullscreen in AMD Radeon Software or NVIDIA Control Panel.

Mouse, Windows, and Raw Input Settings

Raw Input (m_rawinput 1) is the single most impactful aim command. It routes mouse movement data directly into CS2, bypassing all Windows mouse processing. Without it, CS2 applies Windows sensitivity scaling on top of in-game sensitivity — any misconfiguration in Windows multiplies through your aim. Confirm yours: open console, type m_rawinput. If it returns 0, fix it: m_rawinput 1 then host_writeconfig.

Mouse Acceleration (m_customaccel 0) must be off. Acceleration changes your effective sensitivity based on how fast you move the mouse — a slow tracking sweep feels different from a fast flick shot. This makes building consistent muscle memory structurally impossible: there’s no stable 1:1 relationship between physical movement and crosshair movement to encode. Near 100% of CS2 professionals disable this.

Enhance Pointer Precision is Windows’ hidden acceleration layer. With raw input active, CS2 ignores it — but turn it off anyway. Some games don’t use raw input by default, and hidden acceleration in your system configuration causes inconsistent aim the moment you switch games.

Windows Pointer Speed: set to position 6 of 11 (the centre notch). This applies a 1:1 Windows pointer multiplier with no hidden scaling. At any other position, Windows modifies raw movement values in ways that can affect games without native raw input support.

For keybind setup and movement controls, see our CS2 keybinds guide.

Polling Rate: 1000Hz vs 4000Hz

Polling rate determines how often your mouse reports its position to CS2 per second. At 1000Hz that’s every 1ms. At 4000Hz it’s every 0.25ms.

The practical improvement for aim is real but marginal. A 0.75ms difference sits well within human reaction time variance (typically ±15–50ms) and is imperceptible during normal play. If you own a 4000Hz mouse (certain Razer Viper models, some Logitech G Pro variants), use it at 4000Hz — there’s no downside on modern hardware.

One exception: on hardware below an i5 or Ryzen 5 (pre-2019 generation), 4000Hz polling consumes measurable CPU interrupt processing time and can introduce frame-time spikes. In that case, 1000Hz is correct. Stable frame pacing does more for aim consistency than a sub-millisecond polling improvement.

Which eDPI Range Fits Your Playstyle?

How to Find Your Sensitivity Without Copying a Pro

Copying pro settings is the wrong starting point. A pro running 708 eDPI has 10,000+ hours of muscle memory behind that number. Copying it gives you their setting but none of their calibration. Use this process instead:

1. Start at 800 eDPI. If you have CS:GO or Valorant experience, start at your prior eDPI — sensitivity memory transfers across tactical shooters more than most players expect.
2. Play five deathmatch rounds without adjusting anything. You need baseline observation data, not in-session hope adjustments.
3. Diagnose after those five rounds: consistently overshooting targets (flicking past heads)? Reduce eDPI by 50. Never completing fast flicks in time? Increase by 50. One direction per session only.
4. Commit to any change for three full play sessions (45+ minutes each) before evaluating again. Your motor cortex needs roughly 2,000 repetitions to encode a new movement pattern. Switching after 20 minutes resets that clock.
5. Validate with an aim trainer. Aimlabs’ Gridshot benchmarks flick accuracy. KovaaK’s Clicking scenarios test tracking precision. Record a baseline score, change eDPI by 50, play three sessions, compare. If your score improved, continue in that direction.

One rule that applies at every eDPI: don’t change sensitivity after a single bad session. Inconsistent aim is more often explained by fatigue or tilt than by a wrong number.

Your crosshair works with your sensitivity, not independently — see our CS2 Crosshair Guide for settings that complement any eDPI. Spray control interacts with raw sensitivity differently from click-aim; the CS2 spray pattern reference covers the recoil mechanics in detail.

Launch Options and Config File

All aim settings should live in an autoexec.cfg so they persist across updates and reinstalls.

File location: Steam/steamapps/common/Counter-Strike Global Offensive/game/csgo/cfg/

Create autoexec.cfg containing at minimum:

m_rawinput 1
m_customaccel 0
sensitivity [YOUR VALUE]
zoom_sensitivity_ratio_mouse 1.0
host_writeconfig

Add +exec autoexec to your Steam launch options: right-click CS2 in the Steam library → Properties → Launch Options.

Recommended launch options for 2026:

+fps_max 0 -nojoy -high +exec autoexec

Avoid the old CS:GO mouse commands (-noforcemaccel, -noforcemparms) — they have no effect in CS2’s Source 2 engine. -novid is also deprecated; CS2 skips the intro video automatically.

Frequently Asked Questions

Should I copy my favourite pro’s exact sensitivity?
Use their eDPI as a range reference, not a target number to copy. Their setting has thousands of hours of calibrated muscle memory behind it. Find which role-range matches your playstyle from the table above, start near the middle of that range, and calibrate from there.

Is 4:3 stretched resolution allowed in competitive play?
Yes. Valve permits it, professional tournaments allow it, and it has been standard in competitive CS for over a decade. It’s a genuine trade-off (wider models vs. narrower FOV), not a cheat.

Does higher polling rate improve aim?
Marginally, and only after everything else is correct. Fix raw input, disable acceleration, and stabilise your eDPI first. The gap between 1000Hz and 4000Hz is a sub-millisecond improvement — the last thing to optimise, not the first.

How long does it take to adjust to a new sensitivity?
Budget two full weeks of regular play. The first 48–72 hours feel actively worse as your arm fights old movement patterns. By day 10, the new eDPI typically starts feeling natural. If it still feels wrong after two weeks, it may genuinely not fit your style — shift by 50 in either direction and repeat the process.

What is zoom_sensitivity_ratio_mouse?
This multiplies your base sensitivity while scoped with an AWP or Scout. At 1.0, scoped movement exactly matches your unscoped eDPI. Most AWPers prefer values between 0.82 and 1.0 — lower means the scope moves more slowly for precise long-range shots. Start at 1.0 and reduce in 0.05 increments if your scope flicks feel too fast.

Sources

• CS2 Pro Settings and Gear List — ProSettings.net
• s1mple CS2 Settings & Equipment — Specs.gg
• CS2 Pro Settings 2026 — Pley.gg
• Best CS2 Settings for Max FPS & Competitive Advantages — Dexerto
• CS2 Mouse Sensitivity: Best Settings & Conversion Guide — cs2.ad
• Best CS2 Launch Options Guide — Total CS

Michael R.

I've been playing video games for over 20 years, spanning everything from early PC titles to modern open-world games. I started Switchblade Gaming to publish the kind of accurate, well-researched guides I always wanted to find — built on primary sources, tested in-game, and kept up to date after patches. I currently focus on Minecraft and Pokémon GO.