Straight answers to the questions people ask most. If something here isn't clear, everything is explained in more depth across the site. For setting up Clara Core, the Setup Guide has everything you need.
Yes. We explain why it's free on the Why Free page.
Yes. FastPass is a separate tool for passing prop firm evaluations, free like Clara Core: no subscription, no paid tier, no credit card. What it is, and the risk it carries, is under Setup & Use below.
The Micro E-mini Nasdaq (MNQ) futures contract, during the New York session.
At most two trades a day, and it stops for the day after the first win. Any trade still open is closed on the first bar close after 3:30pm ET, so nothing is held overnight. The Strategy page covers the day-by-day rules.
No. The strategy runs its own calendar: it skips FOMC days, market half-days, the day after a half-day, and rollover week, one week each in March, June, September, and December. There is nothing for you to watch or decide; on a skip day it posts a notice and does not trade.
Clara Core runs on NinjaTrader Desktop. If you don't already have an account, open one for free using this link: NinjaTrader (referral link). Sign in once to confirm the account is active, then download NinjaTrader Desktop from your account and sign in there too. Desktop runs on Windows; the next question covers Macs.
Not directly. As of August 2026, NinjaTrader Desktop cannot be installed on macOS, and NinjaTrader has no plans to change that. NinjaTrader Web runs in a browser on a Mac, but it does not run automated strategies, so it cannot run Clara Core.
To run Clara Core you need Windows, and there are two ways to get there on a Mac. Run Windows on the Mac itself with a virtualization tool: NinjaTrader mentions Parallels, which is paid; VMware Fusion and UTM are free tools we found online. NinjaTrader does not support any of them directly, and we run Windows, so we have not tested them. Or run Clara Core on a Windows VPS and connect to it from your Mac.
The difference matters. A virtual machine runs on your own Mac, using your own internet and your own power. If your connection drops, the power goes out, or the Mac goes to sleep, the strategy stops managing the trade. A VPS is a Windows machine in a data center with its own connection and backup power, and it keeps running when your Mac is off. For backtesting and learning the platform, a free virtual machine like VMware Fusion or UTM is enough. For running live, on a prop firm or cash account, we use a VPS: Hyonix (referral link), tested daily.
If we learn of a Mac setup that works well, we will say so here.
No. Clara Core is friendly to someone just starting out with trading or automated trading, but there is a learning curve: prop firms, risk management, and how the strategy behaves are all covered on this website, and how to use the NinjaTrader platform is in the Setup Guide.
A prop firm evaluation, including activation fees, runs $85 to $200 per account, and gives you $2,000 to $2,500 of drawdown room on a 50K account. For a personal cash account or a Roth IRA, we suggest $5,000 and up. A VPS, if you want one, is roughly $6.50 to $24 a month. The Structure page covers each account type in detail.
No. There is no support desk, no calls, no 1-on-1, and we do not answer direct messages. Everything you need is on this site, and the Setup Guide covers every step in writing. For anything about Clara Core itself, this site and the guide are the reference. Clara Core gets its yearly build and the data refresh; the new work goes into what comes next, and the Philosophy page says where that stands.
Yes. Discord and Telegram are read-only: they carry the trade notices and nothing else. Instagram and X are where you can comment, on any post, and we read the comments. We will not get to every question, but we watch what keeps coming up, and it changes this site: a question added to this FAQ, a step made clearer in the guide, a number explained better. Strategy requests are a different thing. We build what we would run ourselves, and nothing else.
Nearly everything you need regarding NinjaTrader, related to Clara Core, is covered step by step in the Setup Guide, so start there. If a technical issue persists, a backtest that will not run, NinjaTrader crashing, an unhandled exception warning, no connection, that is NinjaTrader's territory and their support is quick: in our experience the support assistant answers within minutes, and when it cannot solve something, a human agent has come back to us in under a day. Start with Help inside the platform, or NinjaTrader support.
Yes. We test every strategy for weeks and months, in backtests and on our own accounts, and we run the same version ourselves every trading day. Code can still fail in a way testing did not catch, and we would rather tell you that here than have you find out on your own. When it happens we say so, on Discord and Telegram and by email to everyone who downloaded, with what broke, what it affected, and when the corrected version ships. This applies to everything we release: the strategies, FastPass, and the alert scripts alike. Example: on September 8, 2026, the chart correctly labelled the day as a skip day, Clara Core correctly did not trade, and the notice failed to post to Discord and Telegram. Only the message broke; no trade was affected. The strategy itself gets the hardest test every time anything changes: the backtest has to reproduce to the cent before a version goes out.
Yes. Automated notices post to our Discord and Telegram the moment the system enters or exits, along with the daily skip notices, as proof the system trades in real time and a reference to check your own copy against. They are not signals to follow: Clara Core acts on bar close, and by the time a notice posts, the fill has already happened.
Three ways. The quickest is to follow our Discord and Telegram: an automated notice posts the moment the system enters or exits, and on days it does not trade a notice explains why, a calendar skip, the daily conditions, or simply no signal. Your own chart says the same: the strategy labels every day, so whether it traded or skipped is visible without opening anything else. And if you want to check a past day, run a backtest over it; every trade and every skip day shows up there.
Through the email you provide at download and through our social channels. We don't send promotional emails.
Each version runs for its year. Clara Core 26 takes its last trade on December 31, 2026; from December 1 a notice appears on your chart reminding you to update. After expiry it opens no new positions, but any trade still open is managed and closed normally, so nothing is abandoned. Clara Core 27 releases free in late November, and you install it the same way.
Yes. This is high-variance futures trading. You can lose money, and you can lose an evaluation or funded account. Never run it with money you can't afford to lose. Read more about risk and expectations on the Performance page.
Because that is the shape of this strategy: it wins often and loses big when it loses. In the backtest on the $300 model the average winning trade made $126.74 and the average losing trade lost $197.46, about 1.56 times as much, and the strategy came out ahead only because it won 71.45% of the time. A losing run hurts more than the win rate suggests, and that is the variance the rest of this section is about. The Strategy page explains why the strategy depends on its win rate.
We have no way of knowing what future performance will look like. It did not in the backtest: 12 of the 85 months from May 2019 to June 2026 closed negative on the $300 model, and the worst of them, April 2020, closed at -$870.
The honest mechanics first: every copy of Clara Core acts at the same moments, so if enough contracts ever fired at once, the effect would show up as slippage, slightly worse fills at shared entries and exits. Not as the strategy suddenly failing. MNQ is one of the most liquid futures markets in the world, and Clara Core trades it during the most liquid hours of the New York session, so it takes very large simultaneous size before that effect is even measurable.
There is also a structural buffer: most people running automated strategies are running them on prop firm accounts, and it is where most companies selling a strategy point you. Prop firm evaluation accounts are simulated trading, and at many firms funded accounts are simulated too, not live trading with live orders. Orders on a simulated account never reach the live market, so a large share of running copies leaves no footprint on price at all.
A drawdown past the Monte Carlo worst case, about -$3,706 on the $300 model. That is our tripwire, fixed in advance and published. Crossing it is a trigger to stop adding risk and investigate, not proof the edge is dead. Two things can push a strategy past that line: variance, a losing run that a working edge still produces from time to time, or edge decay, the market changing in a way that makes the edge itself weaker. We tell them apart by measuring recent behavior against the published record: the win rate, the size of the losses, and how long recovery is taking. If they still look like the record, it is variance. If they stop looking like it, we stop running it and say so, here and on our channels.
Clara Core is built to run on its own, and we run ours unattended; watching it live invites anxiety and interference.
Your protective order rests at the broker's servers, so an existing stop does not vanish when your machine drops; you can confirm it from your broker's phone app with NinjaTrader fully closed. After a short drop, the platform reconnects and the strategy continues on its own. After a longer outage the strategy may be disabled, and the Setup Guide walks through the exact steps on what to do, along with the daily routine that refreshes the charts before each session. The Environment section covers prevention, including VPS usage.
No, but we suggest one for anyone running live, on a prop firm or cash account: a VPS means the strategy never depends on your home internet or your computer staying awake. We use Hyonix (referral link), tested daily. The Environment section covers the setup.
Our own practice, not a requirement: we update only at the end of the trading week. If a release lands midweek, we run the current version through Friday. After Friday's close: disconnect the accounts, save everything, close NinjaTrader, reopen it so it prompts the update, update, save, and shut down. On the weekend we also restart the VPS with a normal Windows restart, and log back in Monday on the new version.
Two inputs: your risk per trade and a time zone offset. Everything else is locked, including the skip day calendar. If you run a risk value other than the published models, backtest it first and compare it against the defaults; the defaults are the published record.
Clara Core has dozens of internal inputs, toggles, and parameters that define its logic, and they are locked after seven years of testing. This removes the temptation to interfere, second-guess, or over-optimize based on short-term results. What you control is risk sizing and a few operational settings; everything else runs automatically. The Strategy page covers what is locked and what you control.
Yes. Different firms have different rules on automation, so always confirm what your firm permits before you begin. Clara Core closes all positions before market close, which helps with rules that require flat-by-close. We have run Clara Core on Apex, Bulenox, and Take Profit Trader accounts without issues. We have not lost an account from running Clara Core at any of these firms, but we cannot guarantee that will not change in the future.
Yes. A free compliance template, for anyone whose prop firm asks about their trading approach, is available on the Downloads & Resources page. It comes as a PDF and an editable text file; you fill in your own details and describe your own operation.
Yes: prop firm evaluations and prop firm funded accounts of any size, personal cash accounts, Roth IRA accounts, and free NinjaTrader simulation accounts, all at the same time. But remember that more accounts means more total exposure. Scale with responsibility; the Structure page explains when to scale.
Yes. Clara Core does not care what kind of account it runs in. A Roth IRA needs a self-directed custodian that allows futures and a futures account in the IRA's name; what changes inside an IRA, what the process looks like, and how Clara Core fits are on Trading Futures In A Roth IRA. Information, not tax advice.
No. Clara Core is licensed for personal use only. You may not resell, redistribute, decompile, or share the strategy file or its code. What you can share, freely, is the website and our socials: anyone you send here can download the strategy themselves, free.
No. Only download Clara Core from clarasystems.io. Anyone can put our name on a file, and a copy from any other source can be altered or carry malware. The locked file protects the strategy logic, but we cannot vouch for a file that never came from us.
Yes, for the tools we use ourselves: NinjaTrader, Hyonix VPS, and Equity Trust, all on the Downloads & Resources page. Using them costs you nothing extra and supports the work. We do not publish prop firm referral links or discount codes; the next question explains why.
The main one: affiliate agreements commonly restrict what the affiliate can say publicly about the company, and we would rather keep the ability to tell you when a rule works against the trader.
If you decide to use prop firms, use the firm's own public discount code. Apex, Bulenox and Take Profit Trader all publish current codes on their own front pages. The full reasoning, including a caution about using third-party codes, is on the Structure page.
FastPass is a free tool to pass prop firm evaluations in one day, separate from Clara Core. It runs on two evaluation accounts at once: at a time you set, one account enters long and the other short, each with a take profit and a stop loss. If the market moves cleanly in one direction, one account hits the profit target and passes its evaluation, while the other fails. Getting funded fast is the objective, and it carries real risk and real cost you need to understand first. It is for evaluation accounts only. Read the instruction guide, and test it on simulation accounts, before running it on a paid evaluation. Read the full breakdown on the Structure page.
Yes, for free. With a free NinjaTrader account you can download the historical MNQ data and run the backtest yourself in a few minutes, no paid data subscription required. It is simply how you confirm the numbers instead of taking our word. If you are weighing any strategy, ours or anyone's, the ten questions to ask are on How To Verify A Trading Strategy.
Low drawdown first, then win rate and profit factor. Every filter was tested against the full seven-year backtest, and it had to change the result across a meaningful share of the 1,513 trades; a filter that helps on a handful of trades is noise and was not used. When a filter improved one number at the cost of another, we chose the lower drawdown. There are no random skip days: FOMC days, half-days and the session after, and rollover week each showed a lower drawdown and a better win rate across a meaningful sample. The specifics of each filter are proprietary. The filters were chosen against the same seven years the backtest reports, so the result is not a test on unseen data; that is why the Monte Carlo and its buffer exist, and why no backtest or Monte Carlo tells you what happens next. The Strategy page has the longer version.
Because it is the same platform you run the strategy on, and the same one you can verify it in. NinjaTrader exports the full trade-by-trade log, and its free backtester reaches back to MNQ's first trading day in May 2019, so the published window and the window you can reproduce are identical. We run the same logic in TradingView as an independent cross-check, and show both on the performance page.
No. What we publish is the record you can check without trusting us: every trade posted as it happens on Discord and Telegram, a backtest you can reproduce on your own machine, and the strategy itself, free to download and run. None of it asks you to take our word for anything.