Two traders take the same setup in the same stock within a second of each other. One is filled at $4.62. The other is filled at $4.68, and by the time their order completes the stock has moved on. Same idea, same timing, same screen. The difference is what happened between clicking the button and the trade printing, and most traders have never looked at it.
Order routing is the least examined variable in retail trading. It is also, for anyone taking more than a handful of trades a day in fast-moving stocks, one of the larger recurring costs — larger than commissions in many cases, and entirely invisible unless you go looking.
What happens after you click buy
Your order does not go to "the market." There is no single market. US equities trade across more than a dozen exchanges plus a set of off-exchange venues, and something has to decide where your order goes.
That decision is made either by your broker or by you, and which of those is true is the single most important thing to know about your broker.
Payment for order flow, fairly described
Most zero-commission retail brokers sell their order flow. Rather than sending your order to an exchange, they route it to a wholesale market maker, who pays the broker for the privilege of handling it.
The reason this is legal and not straightforwardly predatory is that the wholesaler is generally obliged to match or beat the national best bid and offer, and frequently fills orders inside the spread. That is real price improvement, and for a retail investor buying 100 shares of a liquid large-cap it often produces a better result than sending the order to an exchange. Combined with zero commissions, the arrangement is genuinely good for that user.
Critics make a fair counterpoint: the price improvement passed to the customer is a fraction of what the wholesaler captures from the spread, and the broker's incentive is to route where payment is highest rather than where execution is best. Both things can be true at once — the fill can be better than the exchange alternative and still worse than what a differently structured arrangement would produce.
PFOF is not a scam. It is a trade: you give up control of routing and speed in exchange for zero commissions and some price improvement. That trade is good for investors and poor for active day traders, which is why the argument never resolves — both sides are describing different users.
Where the trade stops working
Four things break for active traders specifically.
Speed. An extra routing hop costs milliseconds. Irrelevant if you hold for six months. Very relevant in a low-float stock moving thirty cents in two seconds, where the difference decides whether you are filled at your price or chasing.
No route control. You cannot direct the order. In a fast market, the ability to send directly to the venue where liquidity actually is — rather than to a wholesaler's internal system — is the difference between a fill and a miss.
Thin stocks. Price improvement inside a one-cent spread is worth very little. In a small cap with a fifteen-cent spread and fragmented liquidity, where your order goes matters enormously, and this is exactly where wholesaler handling is least advantageous.
Shorting. PFOF brokers typically have limited or no locate inventory for hard-to-borrow names. If you short small caps, this alone usually decides the broker question, independent of everything else here.
Direct market access and route selection
A direct market access broker lets you choose the destination. Instead of one buy button, you have a route selector: ARCA, NSDQ, EDGX, BATS, IEX, various smart routes, and typically a set of configurable algorithmic routes.
Choosing a specific venue sends the order there and nowhere else. It arrives fast and it executes only against liquidity at that venue. A smart route scans multiple venues and distributes the order, which is more thorough and slightly slower.
The practical division most active traders settle on: direct routes when speed matters and you can see where the liquidity is, smart routes when you want completeness and the stock is liquid enough that a few milliseconds are not decisive.
Maker-taker, and why some routes pay you
Most exchanges operate a maker-taker fee model. Adding liquidity — posting an order that rests in the book — earns a small rebate. Removing liquidity — hitting a bid or lifting an offer — incurs a fee. Both are fractions of a cent per share.
At one hundred shares this is rounding error. At several thousand shares, dozens of times a day, it becomes a real line in your P&L. A trader who habitually takes liquidity pays this every time; a trader who posts and waits collects it. Neither is right, but a trader who has never noticed the fee structure is paying for a choice they did not know they were making.
What actually protects you
Regulation NMS includes an order protection rule preventing trades from executing at prices worse than the best displayed quote available elsewhere. That is a genuine floor and it applies regardless of routing.
What it does not guarantee is worth being precise about. It protects against executing outside the national best bid and offer. It does not guarantee you got the best available fill, does not protect against slow routing while the market moves, and says nothing about size — the protected quote may be for 100 shares while your order is for 2,000.
The floor is real and it is a floor, not a standard of excellence.
The disclosures nobody reads
Two SEC rules require the data to answer these questions, and both are public.
Rule 606 requires brokers to publish quarterly reports on where they route customer orders, broken down by venue, including whether they receive payment for that flow and how much. If you want to know whether your broker sells your order flow and to whom, this report says so directly.
Rule 605 requires market centers to publish execution quality statistics — effective spreads, price improvement rates, speed.
These are dry documents and reading your own broker's 606 report once is worth more than any amount of forum argument about which broker is best. It takes about ten minutes and it is specific to the firm holding your money.
Counting the real cost
Broker comparisons usually stop at commission, which is the least useful number for an active trader. The full cost of a round trip includes commission, ECN and routing fees, the spread you crossed, slippage against your intended price, borrow costs if short, and platform or data fees.
A zero-commission broker with slow routing that costs you two cents of slippage per share on 2,000 shares has cost you $40 on a trade a direct-access broker would have charged perhaps $2 to execute. The zero-commission option is dramatically more expensive and the invoice never shows it.
This is the calculation to run before switching brokers, and it depends entirely on your own size and frequency, which is why generic broker rankings answer the wrong question.
Measuring your own execution quality
All of the above is theory until you measure your own fills, and your journal is where that happens.
The core field is intended price versus actual fill. Record the price at which you decided to act, then the price you received. The difference is your slippage, and it is the number that makes routing concrete. Most traders have a strong opinion about their execution and no data.
Record the route or order type used, so different choices can be compared. Record share size, because slippage scales with it and an average across all sizes hides the problem. And record the stock's liquidity, since the entire argument predicts that routing matters more in thin names — a prediction you can test against your own trades.
After a few weeks this answers questions no broker review can: what your average slippage is in cents and dollars, whether it is worse in thin stocks, whether it worsens with size, and what it costs you per month. That last figure is what makes the broker decision obvious in either direction. It is the same approach as any other structural cost, and it belongs in the same review cadence as tracking trading mistakes, because unlike a mistake it happens on every single trade.
A worked example
A trader takes roughly 12 trades a day, averaging 1,500 shares, on a zero-commission broker. They log intended price against fill for a month.
Average slippage: 1.8 cents per share on entries, 2.4 cents on exits. Round trip that is 4.2 cents on 1,500 shares, about $63 per trade. Across 12 trades a day, roughly $756 daily, or somewhere near $15,000 a month in execution cost on a nominally free platform.
A direct-access alternative charging $0.004 per share would cost about $12 per round trip in commission. Even if it only halved the slippage, the trader would be several thousand dollars a month better off.
The numbers are illustrative, but the shape is the point: for an active trader, execution cost routinely dwarfs commission, and it is only visible if intended price is recorded alongside the fill. A trader taking two trades a week would reach the opposite conclusion from the same exercise, which is exactly why it has to be your own data.
Common mistakes
Comparing brokers on commission alone. The largest cost is usually not on the invoice.
Not recording intended price. Without it, slippage cannot be computed at all, and this is the single field most journals omit.
Assuming zero commission means low cost. It means the cost has moved somewhere you are not looking.
Switching to direct access before you need it. DMA platforms carry data fees, platform fees, and per-share commissions that make them more expensive for low-frequency traders. The switch should be justified by your own slippage numbers, not by the fact that professionals use it.
Blaming routing for chasing. Entering late at a worse price is an execution decision, not a routing failure. Recording intended price is also what separates these two, which is part of why it matters.
FAQ
Is payment for order flow bad for me?
It depends entirely on how you trade. For an investor buying liquid stocks occasionally, PFOF plus zero commission is usually a good deal and often delivers price improvement over the displayed quote. For an active day trader in fast or thin stocks, the loss of routing control and speed typically costs more than the commissions saved.
How do I find out if my broker sells my order flow?
Read their Rule 606 report, which they are required to publish quarterly. It names the venues they route to and discloses payment received.
What is direct market access?
A broker arrangement where you select the venue your order is sent to rather than the broker deciding. It is standard on professional day trading platforms and generally comes with per-share commissions instead of zero-commission pricing.
Which route should I use?
There is no universal answer, and it depends on the stock, the moment, and whether you are adding or removing liquidity. The general principle is direct routing when speed matters, smart routing when completeness matters.
Does routing matter if I only trade large caps?
Much less. Spreads are typically a cent, liquidity is deep, and price improvement from a wholesaler is often genuinely competitive. Routing matters most where spreads are wide and liquidity is fragmented.
Conclusion
Routing is the part of trading that happens after the decision, which is why it gets so little attention and why it quietly compounds. The right broker structure is not a matter of principle; it is a function of how often you trade, how large, and in what kind of stock.
That question has a numeric answer specific to you, and it comes from one field most traders never log: the price you intended versus the price you got. Record it for a month and the decision makes itself.
TheSpeculatorsJournal lets you record planned and actual prices on every trade along with custom tags for route or order type, so slippage can be measured per trade and reviewed by stock liquidity and size. You can try it free for 7 days — Basic is $19/month and Pro is $29/month after that.
This article is for educational purposes only and is not financial advice. TheSpeculatorsJournal does not recommend specific securities, brokers, entries, or exits, and past performance of any setup does not guarantee future results. Always do your own research and consider consulting a licensed financial professional before making trading decisions.