Ba Bem

Documenting my daily grind in Trading & Web3, personal perspectives and reflections on the journey to Financial Freedom.
#BaBemTrading

[Ba Bem @BaBemTrading]

Day 1:
Hello world!
Inspired by DiDi, I created this account to document my daily grind. Nothing fancy, just keeping it simple. As someone who often works solo, I frequently have ideas, thoughts, and emotions that I want to share, so I’m putting them here.
Just think of this as stumbling upon a stranger’s diary, that’s all.

[Ba Bem @BaBemTrading]

Day 35:

Hello.
First, here is a brief recap of my work over the past two weeks. I’ll keep it as concise as possible.

1. I ran my 2 main bots with higher and more realistic trading volumes. After 2 weeks, both bots blew up, costing me around $1,700 in total. The returned data was brutal. I had done 3 prior tests with 2 positive outcomes and 1 break-even. However, this time was a complete failure.
- The Reason: I reviewed everything thoroughly and accepted that this was simply a specific market regime; the failure was unavoidable.
- The Solution: * This is the critical part: I realized my monthly ROI was too low compared to the RISK I had to take. Continuous losses would erode my capital fast, while the expected REWARD wasn't high enough >> I’ve decided to completely drop all expectations for these 2 bots. The underlying logic of these 2 bots was user trading psychology. I used to trade this logic manually with great success, but applying it to an automated bot just isn't cutting it. It's just not enough.
I've added σ (standard deviation) and calculated gaps alongside AVGgap5-25 and MEDgap15-30 values as baseline references for the next test. Cutting volume to the absolute minimum, I will return to testing with dust sizes and zero expectations just to validate the logic.
Upcoming bot framework: User Psychology + Price Volatility.

2. I deployed 2 more VPS and built 10 additional sub-bots based on the core logic of the first 2, scaling up both risk and expected reward. However, after just a single week, all 10 were wiped out. I wasn't surprised by the failure itself, as I needed the raw data to optimize. But failing this early? Honestly, I didn't see that coming, haha.
Right now, I’ve patched one of the first 2 main bots, which took a full day. Next, I’ll start debugging the second bot and those 10 sub-bots. It will probably take weeks, but it has to be done.

3. During these 2 weeks of running bots, I started diving deep into σ and Fair Value. I am currently running an observer bot for the BTC 5m market, utilizing σ derived from Deribit and EWMA to calculate FVD/E, incorporating bev% for comparison, and calculating the actual σ-Poly.
To be honest, I don’t fully visualize how these will play out yet, but the math is fascinating and it "looks" rock-solid logically. So, I’ll keep tinkering with it.

4. I’m tracking a specific account on Polymarket. His performance is insane, literally a money-printing machine. I’m purely tracking his trading data to reverse-engineer the model he uses for his bot. This is part of my plan to copy trade successful strategies in this market. I think it’s a solid idea, though it’s going to be an uphill battle.

That’s everything I’ve been up to on Polymarket over the past 2 weeks. Overall, things are getting complex with way more variables to monitor, and a ton of tinkering and heavy math ahead. I’ve realized this is just the beginning. I've been vibe-coding for 2 months now, and it seems I’m genuinely falling in love with this craft. Despite the financial drawdown, looking at the bigger picture, I remain highly optimistic.
Tomorrow, I’ll talk about some mental hurdles I faced and how I overcame them. See you guys then. :)

[Ba Bem @BaBemTrading]

Day 34: but its not day 34.

Hello, it’s been a while.

The past two weeks have been pretty draining, both mentally and physically. A lot has happened, and I was on the verge of burnout. I’ve learned a lot, conceptualized new ideas, accepted a few failures, and still chose to keep moving forward on this path.

Even if no one sees this post right now, I will keep sharing. This time, I’ll be more consistent. I’m treating X simply as my personal journal, for my future self to look back on, for anyone who happens to stumble upon it, and for my partners, with no other motives.

It’s the weekend, go touch some grass. Tomorrow, I’ll recap everything that went down over the last two weeks. See you around. ;)

[Ba Bem @BaBemTrading]

Day 33:

Still no explanation and zero compensation from @Polymarket. I’m at a loss for words :)
It’s their playground, after all. It wouldn't surprise me if they pull even weirder stunts in the future, leaving small-scale users like me with no voice.

Well, as someone once said: "Don't cry in a casino," haha.

I’m moving forward with more bot adjustments. I’ve added two additional price-check layers before a trade triggers. Besides monitoring price changes via WebSocket, I’m now cross-referencing bid/ask conditions directly from the order book depth.

If more "tricks" or weird bugs pop up, I’ll have to consider calling APIs from Binance or Chainlink for external price feeds. Things have taken a turn for the worse here. :/

[Ba Bem @BaBemTrading]

Day 32:

Another day, another bug.

I was trading the BTC 15m market today. Throughout the event, BTC price surged well above the initial reference, and during the final 3 minutes, the price hovered at $0.98 - $0.99 with a very deep order book. Suddenly, in the final 2 seconds, my bot received a WebSocket alert that the "UP" price crashed to $0.04. This is completely irrational. Both of my bots immediately triggered stop-loss sell orders.

It might be a WebSocket data glitch or a corrupted tick, but here’s the crazy part: My bots immediately placed a limit sell at $0.15. At that exact moment, the order book was still packed with "UP" bids at $0.99 since there were only 2 seconds left and the outcome was 100% certain. BTC hadn't moved at all.

Both bots placed sells at the same level. One bot sold everything at $0.99 (as it should), while the other got filled at $0.15. I checked the activity logs, every other user was buying and selling at $0.99 at that exact time. I was the only one whose order filled at $0.15.

I honestly cannot fathom how this happened. I’m starting to suspect @Polymarket 's matching engine or order priority logic at that specific moment. 2 seconds left, and this happens. I’ve opened a ticket, but I haven't received a single reasonable explanation.

It’s pathetic. They’ve raised so much capital and have been operating for years, yet they keep hitting these amateur bugs. And their support team? Absolute shjt. I’ve opened multiple Discord tickets in the past, and not a single one has ever been properly addressed.

Extremely disappointed.

[Ba Bem @BaBemTrading]

Day 31:

I had high expectations for my first two bots.

I had good reason to, because of their logic and how they worked. Although stop-loss trades were unavoidable, after testing them for a while, I became more and more confident in them.
However, any system needs a sufficiently long time to prove its logic is effective and for the bot to actually generate profits. This always means that losing streaks can occur at various times, as can winning streaks.
Fortunately, I was already very familiar with stop-loss orders and losing streaks, as I used to be a full-time HTF trader. Experiencing the psychological phases of trading in the past gave me the resilience to face difficulties with the bot now. I feel really lucky :)

[Ba Bem @BaBemTrading]

Day 30:

Back to work after a week offline for family matters.

I’ve been running two bots for the past 8 days, and both are slightly in the red due to the price glitches on @Polymarket. It took me 5 days more just to patch the system after those hiccups.

The issue is temporarily resolved. Here’s the breakdown: When the bot fetches price data from Polymarket, there are two sources, the "ask price" from price change alerts and the "ask side" from the order book depth. Somehow, last week, the "ask price" repeatedly spiked to 0.9x - 1 while the actual "ask side" in the order book remained unchanged. This triggered the bot to enter trades automatically, only to hit an immediate stop-loss once it detected the price discrepancy.

I’ve now switched all data fetching to rely strictly on the order book. I also added another defensive layer: a price-check mechanism that scans for unusual spikes in the preceding seconds to trigger a cooldown if needed.

I’m putting both bots back in the wild today.

One thing is crystal clear: There will never be a "final version" of a bot. It will always require maintenance, modification, and replacement to adapt to the changing states of the market, the platform, and whatever else comes our way.

Every setback is an opportunity to grow.
Keep pushing, everyone! :)

[Ba Bem @BaBemTrading]

Day 21:

I think I've finished my first bot.

All the logic and backtesting are pretty much working.

In the second week of testing, I increased the volume to 4x the initial amount and made a few changes.

The results are also better than before.

Now the only thing left is to optimize the volume to reach the maximum level the market can absorb without changing the bot's original logic.

Today is the first day I've started running the bot fully automatically 24/7. Hopefully, everything goes smoothly.

WAGMI!

[Ba Bem @BaBemTrading]

Day 29:

Nothing has changed today.

In fact, the errors have gotten worse.
Yesterday, there were 8 more similar faulty markets, at least 5 of which had failed two or more times.
I haven't received any explanation from @polymarket either. They haven't even replied to my ticket.
Looks like I'll have to figure it out myself :)
I'll be offline for a few days to fix this bug.

[Ba Bem @BaBemTrading]

Day 28:

For those who think there's no manipulation on Polymarket :) I hope @Polymarket sees this and gives me an explanation.
Today, April 6th, I was trading BTC 15m as usual, using both my bot and manual execution.

In several markets today, while the Orderbook appeared normal, my bot suddenly received WebSocket alerts that the price on one side had spiked to $0.99 - $1. This immediately triggered my bot’s buy order, which then sold instantly after detecting that the actual price didn't match the alert. I observed at least 5 BTC 15m markets where this happened. The Orderbook didn't budge, yet the price triggered at 0.99/1$.

I immediately tested this by creating a bot to place limit buy orders 100 shares at 1 cent on both UP and DOWN sides. The results? Those orders actually triggered, but the transactions failed to go through on-chain and were canceled instantly. This repeated multiple times within a single market: DOWN spiked to $1, then UP spiked to $1, all while the Orderbook showed zero change, BTC price was nearly stagnant, and there was plenty of time left before the market closed.

Okay, maybe it’s an Orderbook bug, low liquidity, etc. (even though I didn't see any sudden liquidity withdrawals). But why, in that exact scenario, did my $0.01 buy orders fail while $0.99 buy orders filled instantly?? My bot actually bought successfully at the 0.5 - 0.6$ range when the actual price was 0.2$, just because it was alerted that the other side hit $0.99 - $1. This is completely irrational.

I’ve recorded the moment I tested the $0.01 limit buy bot, you can see it for yourself. The side-by-side image shows my bot after I added an extra price-check layer; it flagged the sudden, fake price spikes to $0.99 on both UP and DOWN, all within 13 seconds.

I bet quite a few automated buy bots on 9x are experiencing this today.
Please give me an explanation!

[Ba Bem @BaBemTrading]

Day 28:

For those who think there's no manipulation on Polymarket :) I hope @Polymarket sees this and gives me an explanation.
Today, April 6th, I was trading BTC 15m as usual, using both my bot and manual execution.

In several markets today, while the Orderbook appeared normal, my bot suddenly received WebSocket alerts that the price on one side had spiked to $0.99 - $1. This immediately triggered my bot’s buy order, which then sold instantly after detecting that the actual price didn't match the alert. I observed at least 5 BTC 15m markets where this happened. The Orderbook didn't budge, yet the price triggered at 0.99/1$.

I immediately tested this by creating a bot to place limit buy orders 100 shares at 1 cent on both UP and DOWN sides. The results? Those orders actually triggered, but the transactions failed to go through on-chain and were canceled instantly. This repeated multiple times within a single market: DOWN spiked to $1, then UP spiked to $1, all while the Orderbook showed zero change, BTC price was nearly stagnant, and there was plenty of time left before the market closed.

Okay, maybe it’s an Orderbook bug, low liquidity, etc. (even though I didn't see any sudden liquidity withdrawals). But why, in that exact scenario, did my $0.01 buy orders fail while $0.99 buy orders filled instantly?? My bot actually bought successfully at the 0.5 - 0.6$ range when the actual price was 0.2$, just because it was alerted that the other side hit $0.99 - $1. This is completely irrational.

I’ve recorded the moment I tested the $0.01 limit buy bot, you can see it for yourself. The side-by-side image shows my bot after I added an extra price-check layer; it flagged the sudden, fake price spikes to $0.99 on both UP and DOWN, all within 13 seconds.

I bet quite a few automated buy bots on 9x are experiencing this today.
Please give me an explanation!

[Ba Bem @BaBemTrading]

Day 27:

Caught my first L of 2026. Part of the game 🥴

[Ba Bem @BaBemTrading]

Day 26:

The bullish flag pattern I mentioned earlier is currently holding its price.

However, we are at the end of the pattern as it's unclear whether wave 5 is complete.

On the 1H chart, a trendline supporting the price is more clearly visible in this wave 4-5 phase.

The result will likely be available today or tomorrow.

Hopefully, there will be a very quick drop to 63k.

[Ba Bem @BaBemTrading]

If the price unfortunately exceeds 68k1, we will most likely lose this trade.
Hopefully that won't happen :/

[Ba Bem @BaBemTrading]

Day 26:

The bullish flag pattern I mentioned earlier is currently holding its price.

However, we are at the end of the pattern as it's unclear whether wave 5 is complete.

On the 1H chart, a trendline supporting the price is more clearly visible in this wave 4-5 phase.

The result will likely be available today or tomorrow.

Hopefully, there will be a very quick drop to 63k.

[Ba Bem @BaBemTrading]

Day 24:

The current situation is worse than before. Previously, I expected an immediate drop when I entered the trade, but that didn't happen. After the recent BTC rebound, the downtrend formed a bullish flag pattern at the end of the sideways cycle, which is a bad sign. Coupled with the fact that the third bottom of the 4H RSI is gradually forming, it's entirely possible that a triple bottom divergence will form, pulling the price up again for another wave. Looking at altcoins, I can't imagine what kind of drop the price might take at this point; typically, it will rise again for another wave.

However, this is also a position where a flash dump could break the pattern because everything I said above is just risk hedging; it hasn't really formed a clear structure yet.

What's confirmed here is that the 4H chart rejects SonicR, so I will move my stop loss to 69500 ​​and continue holding the trade.

[Ba Bem @BaBemTrading]

Day 26:

My two current bots are trading the BTC 15m market, exclusively.

Today, I considered building a data-tracking bot for PM to analyze price action and price jumps to identify recurring patterns and their probabilities. However, I shut that down pretty quickly. It felt like I was going against my own mindset. Instead of just fumbling around with data, I need a pre-defined concept and a solid foundation first; only then do I use data for execution.

So, I’ve completely scrapped the idea of a tracking bot, at least for the BTC 15m events.

Next, I’m going to try pivoting to ETH, XRP, and SOL. I have about a month’s worth of data, but everything is failing so far. I’m thinking of building a counter-bot that trades exactly the opposite of my initial test logic.

Even though I don't have high expectations for these markets, given the manipulation and uncertainty, I’ll give it a shot anyway.

[Ba Bem @BaBemTrading]

Day 25:

What’s the biggest challenge of working solo?

Isolation. That’s exactly it, and I face it from time to time.

Mental fortitude is something I value above all else. If it falters, everything else collapses, because fundamentally, I have no one standing beside me. I constantly have to remind myself of the meaning and purpose behind what I’m doing, to see that I’m creating long-term value, not just short-term gains. That’s the fuel that keeps me going.

If I could, I’d love to make as many friends as possible, but I’m just not there yet. I believe in destiny. For now, I can still handle my life on my own, but I know the day will come when I’ll need to lean on my connections and share the journey.

[Ba Bem @BaBemTrading]

Day 24:

The current situation is worse than before. Previously, I expected an immediate drop when I entered the trade, but that didn't happen. After the recent BTC rebound, the downtrend formed a bullish flag pattern at the end of the sideways cycle, which is a bad sign. Coupled with the fact that the third bottom of the 4H RSI is gradually forming, it's entirely possible that a triple bottom divergence will form, pulling the price up again for another wave. Looking at altcoins, I can't imagine what kind of drop the price might take at this point; typically, it will rise again for another wave.

However, this is also a position where a flash dump could break the pattern because everything I said above is just risk hedging; it hasn't really formed a clear structure yet.

What's confirmed here is that the 4H chart rejects SonicR, so I will move my stop loss to 69500 ​​and continue holding the trade.

[Ba Bem @BaBemTrading]

Okay. The trend has been broken.
Although there wasn't a pullback/sideways movement as expected, I still have to stick to the principle.
Open a short position here, SL at 70700 - I'm using SonicR to optimize. I'm a wave-based trader, so I don't have a fixed TP until I see signs of the wave ending, or a strong pullback, then I'll exit. Key levels to watch: 61k, 53k.

First trade of 2026.
#BaBem001 #BTC #BaBemTrading

[Ba Bem @BaBemTrading]

Day 23:

I’ve deployed a second bot today.
The logic is fairly similar to my current one, but with a heavier emphasis on market sentiment, which I believe makes it a safer play. The trade-off is a lower ROI because the entries are less aggressive, but that’s fine, as long as the logic holds, it’s worth running.
I don’t have massive performance expectations for this one, but I’m looking for better long-term stability.

Will update the result one week later ;)

[Ba Bem @BaBemTrading]

Day 22:

One day after I said I would start the bot running fully automatically 24/24, my VPS crashed :)

[Ba Bem @BaBemTrading]

Day 21:

I think I've finished my first bot.

All the logic and backtesting are pretty much working.

In the second week of testing, I increased the volume to 4x the initial amount and made a few changes.

The results are also better than before.

Now the only thing left is to optimize the volume to reach the maximum level the market can absorb without changing the bot's original logic.

Today is the first day I've started running the bot fully automatically 24/7. Hopefully, everything goes smoothly.

WAGMI!

[Ba Bem @BaBemTrading]

Day 16:

I’ve temporarily fixed the latency issues. Currently, the taker is around 300ms, and the maker is 80ms.

Stop-loss has been further optimized, though it still needs more time for monitoring.

I’ve scaled the volume to 4x, and honestly, watching the bot run at this scale is pretty nerve-wracking. However, since the logic remains effective for now, scaling up to find the right fit was an inevitable move.

There’s still a long road of upgrades and monitoring ahead, including: auto top-up, preparing a counter-bot for worst-case scenarios, optimizing capital efficiency, balance threshold alerts, and Discord webhooks...

Keep pushing. I’ll update the results next week.

[Ba Bem @BaBemTrading]

I'm exiting this position at BE.
It's no longer safe.
#BaBem002

[Ba Bem @BaBemTrading]

Follow the trend.
I'm placing another short position at the current price.
#BNBUSDT.
This order will be slightly riskier because I anticipate a downtrend will begin here. But that might not be the case. Therefore, the volume will be slightly smaller and the stop loss will be closer.
The take profit will be at the key levels as shown in the image. As before, it's flexible.

#BaBem002 #BNB #BaBemTrading