You chose the platform and sized the thesis. The fill decides whether edge survives contact with the book. Limit and market orders bridge order-book mechanics, expected value after fees, and the price you actually own.
A limit promises size at your price or better; the venue fills when contra liquidity exists. A market promises size now at available prices, walking the ladder or automated market maker curve. Limits harvest spread and enforce patience; markets buy speed when edge dwarfs friction.
Every strategy chapter in this module assumed you can get the price you modeled. This chapter makes that assumption honest. Without fill discipline, passive edge and momentum edge exist only in spreadsheets.
When to prefer each
If spread is tight and depth is many times your size, posting inside the spread often beats lifting. If spread is wide and no catalyst is imminent, paying five cents of tax on a fifty-cent contract can erase a thin edge—limit only. If news just broke and your model jumped while the book is still tight, one market buy can be rational; repeated market clicks are death by slippage.
Closing arbitrage needs paired discipline: both legs deserve limits when possible; market on one leg only if the edge buffer covers worst slip. Before resolution, cancel good-til-cancelled orders that might orphan-fill into dispute weeks.
| Situation | Lean limit | Lean market |
|---|---|---|
| Wide spread, no news | Yes | Rarely |
| Tight spread, huge edge vs mid | Often | Once if urgent |
| Post-headline, model repriced | After pause | Maybe one clip |
| Thin AMM pool | Simulator first | Usually no |
Limits on central limit order books
Read the touch and a few levels down. If edge versus mid is smaller than one-and-a-half times the spread, stop—expected value is noise. Post a bid below mid with size modest relative to visible depth; improve once if time passes and news is flat. Track average fill, not last tick. Cancel/replace loses time priority—use sparingly.
A filled limit at 48.1¢ versus a market lift at 49.2¢ on four thousand shares saves real dollars when your fair value is 52¢—often more than the “alpha” in the idea. Your job on a CLOB is queue discipline: join versus improve, split size, and know when to walk away.
Markets on pools and hybrids
Pure pool buys move along a bonding curve; slippage grows with size. Hybrids may exhaust the book then hit the pool—know which leg your app routed. Illiquid either engine means do not trade; market orders donate edge.
After a CPI print, fair value might jump to 56¢ while you pay 51¢ market—four cents over old mid but five cents under new fair. A limit at 49¢ might never trade. You bought information arrival, not spread capture. That is the news-jump story from market dynamics chapters applied to execution.
Fees, multi-outcome, and hygiene
Venue fee tiers, profit fees on capped platforms, and gas on chain fills belong in the executable price. Categorical slates need limits on each leg; sum of asks can exceed a dollar—check parity first. Parent-child trees spike children without moving parents—do not chase the wrong node.
Split size when your order exceeds a fraction of level-one depth. Leave limits through binary events only if you accept orphan risk; many traders cancel a day ahead. Posting aggressive limits without news can be the mispricing—you are paying to move the crowd.
Two-leg discipline
When entering two related legs, post the less liquid leg first; when it fills, work the liquid leg with buffer. If only one fills, you are directional—hedge rules apply. For passive entries, one limit plus one optional add on a cheaper print beats three market clips.
Common mistakes
“Market because limit is slow” pays spread every entry. Limit exactly at fair value without queue position never fills. Chasing with repeated markets. Ignoring partial fills when logging Kelly inputs. Stacking correlated limits that double-fill one macro theme. Using pool mid on screen while the fill route was the pool leg on a hybrid.
Cancel and event risk
Cancel resting orders before major scheduled releases if you do not want accidental fills in the vacuum after everyone pulls quotes. Good-til-cancelled through resolution is how orphan positions appear in dispute weeks.
Maker versus taker mindset
Posting limits is providing liquidity. If you improve the bid without news, you may be the mispricing. Taker flow when edge is wide and time-sensitive; maker flow when edge is thin and patience is cheap.
Mental exercise
On a live contract, write bid, ask, spread, and touch size before looking at mid. Derive mid yourself. If you cannot, the UI trained you wrong—that habit transfers to every strategy in this module.
VWAP reminder
If you buy three thousand shares into a thin ask ladder, your volume-weighted average price may be two cents above the first ask. Log VWAP in the journal, not best ask. Kelly and expected value use what you paid, not what the UI flashed for one lot.
Pair with passive and momentum
Passive entries should almost always be limits. Momentum may lift once, then manage with limits. Contrarian fades should never lift panic asks. Arb should pair limits unless one leg is escaping. The strategy choice dictates order type defaults.
Key ideas
Executable price drives EV, not headline mid. Limits harvest spread; markets buy urgency once. Hybrids and pools demand knowing which leg filled.
What comes next
Executable prices feed the sizing pipeline.
Resolution week behavior
Near halt, makers pull quotes; limits may not fill and markets slip. Cancel working orders early; decide hold versus exit while depth still exists. Order-type choice matters less than liquidity existence in the final days.
Misconceptions
“Mid is my price” overstates edge. “Limits are for cowards” donates spread. “One market order is fine” becomes five. “The app said filled at mid” may hide pool slippage. Execution literacy is half of realized alpha in thin markets.
Review fills weekly: average saved cents versus mid is a metric worth tracking.
Next: Position Sizing: Managing Bankroll Risk