Bitsgap vs Botsfolio: What Does the Typical Bot Make?

Jay Sharma · Founder, Botsfolio

Published November 14, 2026

Two numbers, both published by Bitsgap.

The first is on its homepage. Next to the words "Average profit 30d return" for the Grid Bot, it shows 11%.

The second is in a research post the company published on August 7, 2026. Of the grid bots that customers closed in the first half of the year, 35.14% closed in profit. The median return was 0.51%.

They don't measure the same thing. The homepage does not explain what it measures. But the distance between them is the most useful lesson a bot buyer can learn. The remarkable part is that Bitsgap itself is the one teaching it. Of the five companies we compare in this series, it is the only one where we found platform-wide win rates and medians at all. This page reads those figures closely and then asks what even an honest summary leaves out. We make Botsfolio, so weigh that. Every source is linked.

What Bitsgap published

Here is the table from Bitsgap's post "What Actually Worked in 2026?", with the columns in its own order.

Read the grid row slowly. About one grid bot in three closed in profit. The typical one, the bot in the middle of the pack, made half of one percent.

Average against median

Why is the average grid return, 2.34%, more than four times the median? Because averages are bent by outliers. A few bots with huge gains lift the average for everyone, the way one billionaire walking into a bar makes the average drinker a multimillionaire.

Bitsgap says this better than its critics could. From the same post, on why it shows both numbers: "collapsing them into one headline number would misrepresent what most traders actually experienced".

A second post, "Do Trading Bots Make Money?", makes the point with a case. It profiles one trader who made a 157.08% bot profit on $11,762.28 in a week, running five leveraged futures bots. Then it brings the reader back down: "A case study shows what's possible. The median shows what's usual." And: "In H1 2026 platform data, COMBO bots averaged +15.77%, while the median closed COMBO bot made +1.76%." Its summary is one any bot company could print on its front page: "Most results are modest."

That is honest writing. It also sits oddly beside a homepage that leads with an 11% average and no explanation of which bots, which 30 days or whether bots closed at a loss are counted. We are not saying the 11% is wrong. We are saying nobody outside Bitsgap can tell what it is.

What a summary can't show

Credit given, here is what stays hidden even in Bitsgap's better numbers. None of this is a criticism of Bitsgap alone. It is the limit of any summary a company writes about its own customers.

Open bots are missing. The table covers bots that were closed. A bot that is still running and deep underwater is not in it. Neither is one still running in profit. A closed-bot table tells you about the bots people stopped.

Costs are not separated. Bitsgap says so directly: "none of these figures isolate execution cost".

The size of losses is missing. A separate Bitsgap post on DCA bots reports that "45.5% of spot DCA bots closed flat or at a loss" and adds that the research does not publish the typical size of those losses. Knowing how often a bot loses matters less than knowing how much.

The figures change from post to post. In April 2026 Bitsgap wrote that "79% of GRID bots closed with positive realized profit". In August it reported 35.14% closing in profit. The posts do not reconcile the two. Our reading is that the wording holds the answer. Realized profit counts completed trades. A closing result also counts what the bot was still holding. That gap is real and it has a name, covered in the next section. But it is our reading. Bitsgap does not say it.

Nobody outside can recompute it. The source line reads "Bitsgap internal platform data." You can't download the trades and check the arithmetic. You are asked to trust a company's description of its own results. In this case the description is unflattering, which is a reason to believe it. It is still a description.

One more figure from Bitsgap's own reporting says something about how these products get used. In its September 2026 numbers, half of the bots launched in the third quarter ran for 1.2 days or less.

Why a bot is green while the balance is down

Bitsgap has an article titled "Why Your Bot Can Be Green While Your Balance Is Down". The title alone tells you how often customers ask.

The answer is the same on every platform. A grid bot's profit counts its completed buy and sell pairs. It leaves out the loss on the crypto the bot bought on the way down and still holds. Bybit's help center gives the cleanest version: "it's possible to have a positive Grid Profit while simultaneously experiencing a negative Total P&L, indicating that the realized profits are not enough to cover the unrealized losses".

We walk through the mechanics, with the exchanges' own words, in Pionex vs Botsfolio.

What about Bitsgap's AI?

AI made building a bot easy. It did not make the idea any better.

Bitsgap has two AI features and, true to form, it describes them more plainly than most. In AI Chat you type a budget, a crypto and a risk level. The AI picks a bot type, fills in the settings and shows a backtest. Then it stops: "The decision to launch is always yours."

Its blog is blunter still about the AI Assistant. It "doesn't trade on its own judgment", the post says. And: "The AI Assistant is a configuration and portfolio layer, not a discretionary trading agent." Another Bitsgap post, on the ways to automate trading, adds the caveat that belongs on every product of this kind: "Watch out for: AI picks settings from past data."

So the AI chooses the settings. The bot it launches is still a grid or a DCA bot. The table at the top of this page is the record of what those bots typically made. A grid set up by AI is still a grid.

Botsfolio starts somewhere else. There is no bot to set up. You pick the cryptos and the amount. Botsfolio AI does the research and testing and trades them on your own exchange account. Every trade it makes is published.

Is Bitsgap safe?

On design, yes. Bitsgap connects to an exchange account that stays yours. Its homepage says: "Bitsgap doesn't have access to funds on your exchange and cannot withdraw them." It also says the platform "automatically rejects any API key with an enabled withdrawal function", which is the right default and one Botsfolio shares.

On history, Bitsgap describes its own record this way: "Since its launch in 2017, the platform has maintained a spotless security record with no hacks or breaches". That is the company's statement. We looked for an independent report of a breach and found none.

What Bitsgap costs

On November 13, 2026 the pricing page listed a free plan and three paid plans at $29, $69 and $149 a month on monthly billing, cheaper if you pay for a year, with a 7-day trial of the top plan.

Put that next to the median. A grid bot that makes 0.51% on a $2,000 position has earned about $10. A month of the cheapest paid plan costs $29. The subscription has to be earned back before the bot has made you anything. The sums for every account size are in what crypto trading bots cost.

What you have to do yourself

Bitsgap's data makes an argument its marketing does not. The spread between the median and the average is the spread between customers. A few of them set their bots up very well. Most did not. The tool was the same.

How Botsfolio is built

Botsfolio does not ask you to be one of the few. You pick the cryptos and the amount. You can stop at any time. Botsfolio AI does the research. It generates trade ideas, tests them on years of market data, finds what is working and uses that to trade the cryptos you picked. Most ideas fail its tests and are never used.

And where Bitsgap publishes a summary, Botsfolio publishes the trades.

Every filled trade is public. Paper and real money, each goes into a log that is sealed every hour, signed by Botsfolio and anchored in Bitcoin. A seal shows who published a row, when it existed and that it has not changed since.

Anyone can redo the arithmetic. The record pages come with downloads. The verify page has the code to check a seal. If our summary and our trades ever disagreed, you could catch it.

Finished periods don't change. Each week and month has its own page, for real money and for paper trading, with the return, the worst drop and what holding the same cryptos did. A bad week stays a bad week.

A caution in the other direction, because it is only fair. These are different kinds of data. Bitsgap's table describes thousands of bots set up by its customers. Botsfolio's record describes what one AI did. Neither is a forecast.

Here are the three records as of the last daily close:

Where Bitsgap is the better choice

• You want to run your own bots. Grid, DCA and futures bots with settings you control. Botsfolio offers none of them and trades spot only.
• You want to practice first with the same tools. Bitsgap has a demo mode and a free plan. Botsfolio's free paper trading lets you watch the AI trade, but not build a bot.
• You value a company that prints its unflattering numbers. Bitsgap's research posts are more candid than anything else we read for this series.
• You want help setting up a bot. AI Chat takes a budget, a crypto and a risk level and prepares the bot for you to launch.

If you do set up a Bitsgap bot, its own data suggests the questions to ask of it after a month. Is the total up, or only the bot profit? Up by more than the subscription? Up by more than holding? Our guide to the buy and hold comparison shows how to answer the last one.

Side by side

FAQ

Sources

Botsfolio is software. It is not investment advice and it is not tailored to you. Botsfolio has no affiliation with Bitsgap. Bitsgap is a trademark of its owner, used here only to name the product. Paper trading results are simulated. Trading crypto can lose money. Past performance does not predict future results.