I kept getting stopped out by walls I couldn't see, so I merged 20 order books into one
For a long time my routine looked like this: Binance order book in one tab, Bybit in another, OKX in a third, and Hyperliquid on my phone because I ran out of screen. I'd spot a big bid on Binance, feel brave, go long, and then watch price slice through it like it wasn't there.
It took me embarrassingly long to understand why. That wall wasn't small. It was just one piece of the market. The same price level had orders on eight other exchanges, and the sellers hitting it were spread across all of them. Looking at one book was like judging traffic on a highway by watching a single lane.
What actually changed when I added the books together
The first time I summed the books properly (same price grid, contract sizes converted, USDT vs USD pairs lined up), a few things jumped out:
- BTC perps had roughly $70–115M per side within 0.1% of price across 11 exchanges. The biggest single venue held maybe a quarter of that. So "the Binance book" was a quarter of the picture at best.
- Walls that looked huge on one exchange were often nothing in the total. And sometimes a level that looked boring everywhere was quietly big because five venues each had a medium order there.
- Some walls appear and vanish without a single trade. On one exchange you'd never notice the pattern. Across all of them it's obvious.
None of this is secret knowledge. It's just tedious to do by hand, every second, for every coin.
The parts that were harder than I expected
Contract sizes are a mess. One exchange quotes PEPE per 1,000 coins, another per coin, a third in contracts with a multiplier hidden in the docs. Get it wrong and your "wall" is 1,000× too big.
Prices differ slightly between venues, so you can't just stack levels by exact price. You need a common grid and a way to decide which bucket each order lands in.
And the data never stops. Dozens of WebSocket connections, reconnects, snapshots that arrive late, exchanges that throttle you for asking too often. Most of the work wasn't the math, it was keeping all of it alive at 3 a.m.
What I use it for now
I mostly look at three things before a trade:
- How much money sits between price and my stop, summed across all exchanges. If my stop is behind a level that's thin everywhere, it's not really protected.
- Where the liquidation clusters are. Price likes to go where forced closes are waiting.
- Whether there's a big wall between my entry and my target. If there is, I either take profit before it or skip the trade.
It didn't turn me into a genius. It did stop me from trusting a wall that was only big in one tab.
_I ended up turning this into a site, AggPro (aggpro.xyz) — the order book, the screener and the coin pages are open without sign-up if you want to poke at it. But honestly, even if you never use it: next time a wall looks massive, check the other exchanges before you lean on it.
_
Top comments (0)