Community › Q&A

so how do i size around thin top of book

GGapHunterMike· Aug 2026 ago answered· 👍 2· 💬 3
read the depth study in research. the one where they walked the recorded ladder to see how much size the quoted price actually holds. suspected this. didn't know it was that thin. the top level is a tiny sliver of what's in ten levels. rest is sitting further out. old job i had a dom in front of me all day. you see the size, you know when to lean on it. now i'm clicking a market order in an app and the number on screen is basically decoration. questions. 1. does the study cover premarket. movers gap and i trade the open. i assume book is worse then but i don't want to assume. 2. if my normal clip is bigger than what the top holds, is splitting into two or three actually better or am i just paying spread twice plus giving the market time to move away. 3. anyone got a rule of thumb they use. like cut size on anything under a certain price, or check the ladder before sending. not asking what to trade. asking how people handle the mechanics.
G
GapHunterMike
17 posts · 0 followers
+ Follow

Related reading

Guides & TutorialsWhat is Level 2 market data, and is it worth paying for? We measured 35 million quotesResearchThe quoted price is only good for about $1,200 to $5,200: order book depth measuredGuides & TutorialsHow to size a day trade, measured from the order bookResearchWhat day trading a US stock really costs: 35 million quotes measured

3 replies

TraderWeTraderWe Team· Aug 2026 ago
On the first question — the measurement window there is regular session quotes, so premarket isn't inside that sample. Generally depth is thinner outside regular hours and spreads wider, but we'd rather not put a figure on it until we've measured it the same way. On the second: splitting doesn't avoid the spread, it just changes how much of the book you sweep per clip. Whether it nets out better depends on how fast the name is moving, which is why we don't publish a rule. You can see the book on the order screen before you send.
BacktestBetty· Aug 2026 ago
Three things. (1) That study is a distribution, not a constant, the range they report is wide across price bands, so a single sizing rule tuned to the middle of it will be wrong at both ends. (2) If you go build a size cap from it, be careful you're not fitting to the exact sample window; execution conditions drift and 264 sessions of one regime is still one regime. (3) Easiest honest test is your own fills, pull your fill prices against the quote at send time and see where the damage actually shows up. Cheaper than theorising and it's your own instrument set.
F
FiveMinFiona· Aug 2026 ago
On the third one. I check the ladder before sending, always, and it takes two seconds. If the top level looks thin relative to what I want, I either cut the clip or use a limit and accept I might not get filled. Missing a fill has never cost me as much as a bad one on a fast mover.
Sign in to reply →
← All q&a