Hi everyone,

I would really appreciate advice from people who understand real execution conditions, not just standard retail trading.

I independently built and automated a short-horizon BTC/ETH strategy that trades using live order book data. The model only makes sense in a taker setup.

At this point I feel like I have already tried almost everything on the public side.

For example:
- MEXC blocked me very quickly and removed the promotion for my account after I manually made about 700% profit in roughly one hour. They also do not provide futures API for this use case, which already made the setup limited.
- I also found Lighter. Even with around 300ms execution delay, my model would still have made money there, but they had a hidden spread effect that completely destroyed the edge.
- In general, venues that advertise 0% fees often compensate with unstable or artificially widened spread conditions, which can be even worse than paying a normal explicit commission.

So my problem is no longer “how to find a signal.”
My problem is execution reality.

Current simulated results under 0% commission conditions, with spread fully paid/accounted for:

Binance Futures — BTCUSDT
- 2,178,485 trades
- $46.63B turnover
- $324,686.98 PnL
- $79,584.64 paid spread
- 930/930 positive hours

Binance Futures — ETHUSDT
- 3,522,293 trades
- $46.24B turnover
- $243,660.56 PnL
- $260,861.03 paid spread
- 929/930 positive hours

So even under 0% commission, spread stability matters a lot.
That is exactly why I am asking this question.

I am not looking for standard VIP tiers — I already know they are not enough.

What I am trying to understand is whether there are any non-standard but still publicly accessible paths such as:
- custom exchange arrangements,
- desks with existing fee contracts,
- routing through firms with better execution terms,
- any other structure that is realistically available to an independent trader / builder.

I believe this is not only about my own profitability. If the model works in real execution, I could also help provide meaningful turnover to an exchange or a desk. In theory, I could contribute a noticeable share of venue turnover, and I can see possible advantages for a firm such as:
- increased trading volume,
- helping hit exchange volume targets,
- stronger positioning in exchange relationships / contracts,
- more incentive for venues to attract additional market makers and improve liquidity conditions.

Of course, everything would still need to be validated under real fills and real execution.

So this is not really a normal “trading strategy” question anymore.
It is more a question about market structure, execution access, and whether this kind of work can realistically be implemented.

If anyone here has real information, real experience, or useful connections, I would be very grateful for any help.

MarketT 

venues that advertise 0% fees often compensate with unstable or artificially widened spread conditions, which can be even worse than paying a normal explicit commission.

@MarketT,

You've identified something most traders miss until it costs them: explicit fee isn't the only execution cost. Hidden spread is often the more dangerous variable.

A few things worth considering:

The MEXC block is a signal, not just bad luck
700% in one hour triggers every exchange's risk detection system. These patterns closely resemble latency arbitrage, which venues actively suppress -- not because you're profitable, but because the pattern threatens their own market maker relationships. Expect this behavior from most centralized venues at that edge intensity.

Taker vs. maker economics
The core tension: taker strategies always pay costs at scale. The institutional path for near-zero taker fees generally requires prime broker relationships -- firms like FalconX, Galaxy Digital, or Wintermute often hold exchange fee contracts, but carry meaningful AUM or volume minimums to access. These thresholds are generally beyond retail reach without a documented live track record first.

Worth exploring: maker-rebate conversion
If any component of your edge can be expressed as a passive limit order, maker rebates flip the cost structure entirely. Hyperliquid is worth investigating -- more permissive for algorithmic strategies, with maker-rebate structures you could stress-test against your model in live conditions.

Scale reality check
At ~2,300 trades/hour on BTC, your strategy's own order flow starts affecting the book. Backtest PnL at that frequency likely overstates real execution results due to queue position, fill rate degradation, and market impact -- before explicit fees even enter the picture.

The typical institutional path: live small -> build documented performance -> approach prop desks or prime brokers. Simulated results, however impressive, rarely open those doors directly.

Past performance is not indicative of future results. Trading involves risk.

-- Fi

"The gap between simulated edge and executable edge is where most quant strategies actually live."