NO CODE REQUIRED

A quant tool that
builds the quant for you.

You don't need to write code, and you don't need a quant background. Pick factors, drag the weights, run a real backtest — and walk away with a strategy that is actually yours.

LIVE ON Binance OKX Bybit Bitget Gate KuCoin MEXC Propr + any exchange with an API
ZHQDSK — Strategy
FACTOR WEIGHTS
Price position
Distance from low
Funding rate
Momentum
Volatility
BACKTEST RESULT t = 4.33
+189.8%
178 periods Mean +0.60% Max DD 10.52%
BTC +\.40 ETH -\.10 SOL +\.22 DOGE +\.05 SUI +\.18 APT -\.44
2:41
Windows
2:41 ▮▮ ▮
ZHQDSK
TODAY
$5,120.44
+2.41% today
Positions56
Sizing$35
Run strategy
BTC+$12.40
ETH-$3.10
SOL+$8.22
Android
Every screen earns its place — Windows and Android, same engine, same numbers. A phone alone is enough to run a quant.
The part nobody posts

Still trading alone —
and reality keeps hitting back?

Chasing a green candle, cutting at the bottom, adding to a loser because you "believe in it", taking profit at the first sign of fear. You don't lack courage. You lack a repeatable rule.

  • Emotion decides — every order is a mood, not a rule
  • No stop discipline — one bad day erases a month
  • Nothing to review — you can't tell luck from skill
A typical manual account
NO RULES
Bought the top
Cut at the bottom
Added to a loser
Illustration, not a real account. The pattern, though, is real — and common.
What a rule does that you can't

Every way you lose has
a rule that fixes it.

The three things that empty a manual account are the same three things a written rule removes. Below is what that actually looks like on a chart.

01

Emotion never casts a vote

A sharp flush is the moment most people sell. Not because the thesis broke — because the screen turned red. A written rule doesn't see red. It sees whether its exit condition was met. Usually it wasn't.

  • You: sell into the panic, then watch it recover from the sidelines
  • The rule: holds until its own exit triggers — and here, it never did
Same crash. Two decisions.
sold here rule still held
You The rule
02

Chop doesn't get to bait you

Sideways price is where most losses are born: every small dip looks like the discount of the year, so you buy it — and it keeps going down. A rule with a defined entry has no opinion about "cheap".

  • You: buy the first dip in a range, get trapped by the real move
  • The rule: no signal, no position — it sits out the chop
Same range. Two reactions.
bought the dip still holding range — no signal
Price Your entries
03

You can tell luck from skill

A rule that ran 178 periods and survived has evidence behind it. A hunch that worked once has a story. Only one of those can be repeated — and only one of them can be tested again tomorrow on new data.

  • The rule: same inputs, same output — every single time
  • The result: a t-stat and a drawdown you can inspect, not a feeling
Same rule. Three runs.
Dataset A identical
Dataset B identical
Dataset C identical
A deterministic algorithm cannot give you a different answer to the same question.
04

Every position is defended

Three hard limits: a daily loss circuit breaker, a drawdown guard and a stop on every single position. If any one of them trips, the engine stops for the day.

  • You: stop out on willpower — and willpower runs out
  • The rule: stop out on a number — numbers don't get tired
Three layers, always on
Daily loss limit
2.50%
Max drawdown
5.00%
Stops placed
56 / 56
When any layer trips, the engine stops for the day. There is no "try again".
Why not just use AI

Same money. Why still write the prompts?

Price position
Distance from low
Funding rate
t = 4.33
This tool
✓ Open it, pick a preset, run
✓ Same input, same output — every time
✓ $149 a year, unlimited runs
✓ t-value and drawdown you can verify
Write me a cross-sectional strategy. Must handle funding, volatility, costs, and backtest…
Here is a momentum-based approach you could try… (numbers unverified)
Just asking an AI
× You write the prompt — and rewrite it every time it drifts
× Different answer every run. Numbers it cannot back up
× $20 a month, billed by usage
× No backtester. No cost model. No reproducibility
An AI subscription runs about $240 a year and still leaves you writing prompts. This is $149 — the rules are already in the engine, running against real history.
On-device backtester

You set the rules. It tells you the truth.

Tunable parameters
8
all open to you
Pool size How many Selection ratio Hold period Trading cost Long / Short Ranking Sizing
Preset strategies
8
pick a start or build your own
Official template Pure momentum Short reversal Funding Low vol Trend Mean reversion Cost-aware

No hidden parameters, no locked switches. Everything you change here changes the next backtest.

This is a testing tool, not a signal service. We never tell you what to buy. You choose the factors and the weights, run them against 1,200 bars of real history, and get an honest number back — including when your idea loses money.

✕We don't send buy/sell signals
✕We don't pick coins for you
✕It never runs without you starting it
✓
Statistically significant
Green verdict. t ≥ 3 and the second half holds up. This is the only outcome that earns a live run.
!
Positive, but not significant
Amber verdict. It made money in the sample, but the evidence is too thin to trust.
✕
It loses money
Red verdict. On this history, the idea would have lost. Better to find that out here than with real capital.

Same engine, three possible verdicts — and it will not only show you the pretty ones.

Factor weights

Drag a slider. That is your strategy.

live
Price position (96)
Where price sits in the last 96 bars
0.50
Distance from low (48)
Gain from the 48-bar low
0.50
Funding rate
Current funding — expensive to short
1.00
Momentum (96)
Return over the last 96 bars
0.00
Short-term reversal (24)
Inverse of the 24-bar move
0.00
Every weight is yours to set. Nothing is locked, nothing is hidden — the same sliders you see here are the ones that run your backtest.
成绩概览
胜率55.6%
最大回撤10.52%
t 值4.33
每一笔都对得上
周期 收益 成本
2024-03 +2.31%0.11%
2024-04 -0.84%0.11%
2024-05 +1.92%0.11%
选一个起点
官方模板12
动量策略6
自定义36
What else is in the box

The rest of the engine.

A backtester is only useful if what you test can actually be executed. Everything below is what turns a tested idea into a running position.

Exchange-agnostic by design

Binance, OKX, Bybit, Bitget, Gate, KuCoin, MEXC and Propr are live. Adding another venue is not a rewrite — it is one more adapter. Any exchange that issues an API key can be plugged in, and your strategy does not change a line.

Sizing that fits your balance

The engine reads your balance, works out how many positions your capital can actually support, and skips pairs whose minimum order size you cannot afford. No "insufficient margin" spam.

Twelve factors, three ways to score

Price position, distance from low, funding, momentum, volatility-adjusted momentum, reversal, ATR, skew, acceleration, BTC correlation, dollar volume. Weight them yourself, then choose z-score, rank or winsorized scoring.

Where we differ

Mainstream tools sell you a strategy.
We sell you the ability to build one.

Most "quant" products hand you a black box: you subscribe, you follow, you hope. ZHQDSK works the other way around — the rule comes from you, and the tool makes it executable, testable and repeatable.

Mainstream quant software
ZHQDSK
What you get
A finished strategy product
A tool that turns your idea into rules
Can you change it
No — or only by writing code
Every weight is a slider you can move
When you have an idea
Wait for their next release
Build and test it yourself in ten minutes
Who watches the screen
You — or their desk
The program, on the schedule you set
When it loses money
You can't tell why
t-stat, drawdown, and every period on record

Two ways in — pick the one that matches you

A

You already have a strategy and discipline

You know when to enter, when to exit, how much to risk — you just don't want to sit at the screen all day. Write those rules into the tool once, backtest them, and let the program execute them on your schedule.

  • Your rule becomes a running position, not a note
  • No screen-watching — you review results, not candles
  • Unlimited backtests to confirm it before money moves
Standard
B

You don't have an idea yet

Start with the twelve free factors and test one combination after another — unlimited runs on Standard will eventually show you what has an edge. Or take the template we already validated and reshape it until it's yours.

  • Standard: unlimited backtests, build it yourself
  • Ultimate: start from a validated template, then change it
  • Either way, you end up understanding your own rules
Standard · Ultimate
The whole flow

Five steps.
No code. One afternoon.

You don't need to write code, and you don't need a quant background. Every step below is a button, a slider or a result screen — and you can go back and re-test any time.

No coding 5 steps Re-test anytime Nothing irreversible
  1. 1

    Read the starter pack

    New to all this? A short guide explains every parameter in the backtester — what it controls, and what happens when you raise or lower it.

    Optional · 15 min
  2. 2

    Set your parameters

    Pick factors, drag the weights, choose the universe size and how long you hold. All sliders, no formulas.

    About 5 min
  3. 3

    Backtest and find your strategy

    Run it against 1,200 bars of real history. Look at the t-stat, the drawdown, and the verdict. Keep what survives.

    Under 1 min per run
  4. 4

    Turn on automation

    Paste an API key with trade permission only. The app reads your balance and works out how many positions you can afford.

    About 2 min
  5. 5

    Run your own quant

    Dry run first to see the exact orders. Then switch on live mode — one confirmation per cycle, stops attached to everything.

    Any time
0 lines of code to write
~25 min from install to first live cycle
Undoable every step, at any time
Security

Your keys and your data never leave your device.

Most trading tools ask you to hand over an API key to their server. We don't have one to hand it to. The app signs requests on your device and sends them straight to your exchange — there is no middle layer, and nothing of ours to breach.

  • API keys are stored only on your device Encrypted in local storage. They are sent to your exchange and to nowhere else — not to us, not to any analytics service, not to any backup.
  • We never hold a copy of your key There is no account system, no cloud sync and no admin panel where your credentials could be read. If we wanted to see them, we would have nowhere to look.
  • Positions, PnL and balances are never uploaded The app reads them from your exchange, shows them to you, and keeps them on the device. No usage analytics, no trade history collection, no telemetry on your holdings.
  • Only the license code is ever transmitted The single outbound request we make carries your license key and a product ID. No key, no positions, no balance — nothing that could identify your account.
  • Withdrawal permission is not required — and not wanted Create the API key with "trade" permission enabled and "withdraw" disabled. The app only needs to place and close orders; it never needs to move funds out.
  • Uninstall and it's gone Because nothing lives on a server, deleting the app deletes your keys and your data. There is no "please contact support to remove your data" step.
This device
KEY
Your exchange
✕ No server of ours in between
What we can see
Your license code
That a license is active
What we cannot see
Your API key or secret
Your positions or balance
Your trade history
Your exchange account
Signed requests go straight from your device to the exchange. There is nothing of ours to leak.
Pricing

Start free. Upgrade when it proves itself.

Three plans. Each one can be paid monthly, quarterly or yearly — the yearly price is always the cheapest.

7 天無理由退款 付款後 7 天內可全額退款,扣除 3 USDT 主網手續費。 寄信到 zhuxi4685@gmail.com 即可申請。
FREE

Test your ideas. Pay nothing.

The complete backtesting engine, free forever. Wire up your own factor weights, pull real market history, and find out whether an idea actually works before a single dollar moves. Ten runs a month is enough to get an answer.

  • 10 backtests per month
  • 12 of 36 factors, fully adjustable
  • Runs locally on your device
  • No strategy templates
  • No live trading
Start for free
免费版需邮件确认后开通,通常两个工作日内
$0/ forever
No card required
10 backtests reset on the 1st of every month.
STANDARD

Connect your exchange. Let it work.

Everything the free plan does, plus the one thing that matters: it can actually trade. Unlimited backtests, live execution on your own account, three layers of risk control. You still confirm every cycle — but the orders place themselves.

  • Unlimited backtests
  • Live execution — you confirm each cycle
  • 24 of 36 factors · all 8 exchanges
  • Adaptive position sizing
  • Three-layer risk control
  • 2 devices
Monthly
$19/ mo
Choose
Quarterly
$49/ qtr
Choose
Yearly BEST VALUE
$149/ yr
Choose
$12.42 / month on the yearly plan — 35% less than paying monthly. Cancel anytime, access runs to the end of the paid period.
ULTIMATE

Hand it the strategy we already tested.

Standard lets you build a strategy. Ultimate hands you one to start from — the exact factor weights we backtested over 178 periods to a t-stat of 4.33, with the full preset library: momentum, reversal and funding carry. Adjust anything, or run it as delivered.

  • Everything in Standard
  • The validated strategy template
  • Full preset library — momentum, reversal, carry
  • All 36 factors, plus every future release
  • Every future exchange integration
  • 5 devices
Monthly
$29/ mo
Choose
Quarterly
$75/ qtr
Choose
Yearly BEST VALUE
$229/ yr
Choose
$19.08 / month on the yearly plan — 34% less than paying monthly. Includes every future update at no extra cost.
Ultimate · the included template

This is what the template actually backtested to.

The template bundled with Ultimate is a fixed, documented set of factor weights — not a black box. Here is its full historical run, produced by the same backtester you get. Costs included. Nothing hidden.

Fixed reference dataset · 195 coins · 4,380 bars · 178 periods
+189.8%
t = 4.33 · 178 periods · fixed dataset
Periods178
Mean / period+0.60%
Win rate55.6%
Worst period−4.50%
Max drawdown10.52%
These are historical simulation results on 195 pairs, after 5.4 bp per-side costs. They are not a forecast. Markets change regime — that is exactly why the app ships a backtester, so you can re-run this yourself on today's data.
Questions

Answered.

怎么联系你们?
任何问题——购买、授权、退款、技术故障——都请寄信到 zhuxi4685@gmail.com。我们会在两个工作日内回复。
Is this financial advice?
No. ZHQDSK is a software tool. It executes rules you configure, on your own account, when you tell it to. It does not manage money, does not provide personalized recommendations, and does not guarantee any result.
Do you have access to my funds?
No. Your API keys are stored only on your device and sent only to your exchange. We never see them, and we recommend keys with trade permission only — never withdrawal.
Will the backtest results repeat?
Not necessarily. Backtests are simulations on historical data. They can overstate real performance due to costs, slippage and regime change. The figures shown are historical, not forecasts.
Which exchanges are supported?
Binance USDⓈ-M futures, OKX perpetual swaps, Bybit USDT perpetuals, and Propr. Contract sizes differ: OKX contracts are larger, so a full 98-position portfolio needs a materially bigger balance there.
Where is it available?
Not available to residents of mainland China, the United States, or any sanctioned jurisdiction. Purchasing from a restricted region voids the license without refund.
What platforms?
Android (APK) and Windows desktop. Both run the same interface and the same engine.
Contact

Talk to me before you buy.

One person builds and runs this. No ticket system, no chatbot — an email that actually gets read.

Before you buy

Not sure it fits your case?

Tell me the strategy you have in mind and the size of your account. I'll tell you honestly whether this tool suits it — including when the answer is no.

After you buy

Something not working?

Send the order number and your license key. That gets it diagnosed in one round instead of three.

zhuxi4685@gmail.com Write an email →

Replies within two business days. Refunds, license resets and technical issues all go to the same address.