Market maker Level 2: who is quoting and what MMIDs do not reveal
An MMID tells you who displayed a quote, not why. Learn which participants quote small caps, what a retail Level 2 leaves out, and why the same refreshing offer appears before breakouts and fades.
Start in plain English
Level 2 shows displayed orders at several prices, often labelled with a short code for whoever is showing them. Those codes are market participant identifiers, usually called MMIDs; FINRA's rules call them market participant symbols (MPIDs). Small-cap traders spend a lot of time watching them: which market maker is "on the offer", which one keeps refreshing, which one is "the ax".
The core idea of this lesson: an MMID tells you who displayed a quote, not why they displayed it, how much else they hold, or what they do next. Exchanges, market makers and wholesalers quote for different reasons, and much of the trading in a small cap does not appear on the book at all. A label on Level 2 is weak evidence of intent. The same refreshing offer shows up before breakouts and before fades.
US Day Trading Fundamentals covers the basics in How to read Level 2. This lesson starts where that one stops: who the names are, and how much weight they can carry.
Two paths from the same start
All prices, sizes and identifiers in this example are Illustrative. Marrowby Therapeutics is fictional, and so are the MMIDs (MMQA, MMQC, ECNB, EXCH1, EXCH2).
The setup at 09:47:30 ET. Marrowby gapped up and printed a high of $5.24 at 09:41. Since 09:46, a market maker labelled MMQA has shown 1,000 shares on the $5.21 offer. That offer traded and reappeared six times; 6,400 shares printed at $5.21. Chat calls MMQA "the ax holding it down". Here is the book at the decision time.
Marrowby Therapeutics (fictional), Level 2 at 09:47:30, ET (Illustrative)
| Side | Price | Size | Source |
|---|---|---|---|
| ask | 5.21 | 1,000 | MMQA |
| ask | 5.21 | 300 | ECNB |
| ask | 5.22 | 2,500 | EXCH1 |
| ask | 5.23 | 800 | MMQC |
| ask | 5.25 | 5,000 | EXCH2 |
| bid | 5.20 | 1,200 | EXCH1 |
| bid | 5.20 | 400 | ECNB |
| bid | 5.19 | 2,000 | MMQC |
| bid | 5.18 | 900 | MMQA |
| bid | 5.15 | 3,000 | EXCH2 |
Identifiers are fictional. Since 09:46:00, MMQA's 1,000-share offer at $5.21 traded and reappeared six times; 6,400 shares printed at $5.21. Day's high so far $5.24 at 09:41. Displayed quotes only.
The plan, written before 09:47:30. Entry at $5.21 with a marketable limit order (ceiling $5.22), filled at $5.21. Stop at $5.16. Target $5.31 (2R). Time stop at 09:58. Planned risk $0.05 per share. Size: $50 ÷ $0.05 = 1,000 shares. Costs: $0.005 per share round trip (Illustrative), plus any slippage.
Path AThe refreshing offer stops
At 09:48:10 MMQA's offer is taken and does not come back. The next offers at $5.22 and $5.23 trade quickly, price clears the $5.24 high, and the $5.31 limit sell fills at 09:52.
Path BThe refreshing offer keeps going
MMQA keeps refreshing for four more minutes. Bids at $5.20 and $5.19 thin out, prints move to the bid, and price slides. The $5.16 stop triggers and fills at $5.15.
Path CThe offer comes and goes
MMQA steps away, returns, steps away again. Price chops between $5.18 and $5.24 without reaching the stop or the target. The 09:58 time stop exits at $5.20.
The review note reads the same for all three: "Entered on a refreshing offer near the high; plan followed; 1R planned; outcome A, B or C." In Path A traders later say "the ax finally lifted"; in Path B, "the ax was selling all along". Both stories fit only after the fact. The rest of this lesson covers who quotes small caps, what your feed leaves out, and why the "ax" idea is hard to test.
Who shows up on a small-cap book
Four kinds of participant appear, or fail to appear, on a typical retail Level 2:
| Participant | Why they quote | What you see |
|---|---|---|
| Exchanges (lit order books) | They display orders sent to them by brokers and traders | The exchange's identifier, aggregating many unrelated orders at each price |
| Registered market makers | They may have quoting obligations and earn the spread; their inventory and hedges are private | Their own identifier, sometimes at the inside, sometimes far away |
| Wholesalers | Retail brokers route most of their customers' marketable orders to them, according to the SEC's June 2026 proposing release | Often little or nothing on the book; their trades show up later on the tape as off-exchange prints |
| Other traders' orders | Routed to an exchange, they appear under the exchange's label, not under the trader's name | Size at a price, with no owner |
So when you see "EXCH1 2,500 at $5.22", that is the sum of everyone who sent an order there. When you see "MMQA 1,000 at $5.21", that is one firm's displayed quote, but not its position, its other quotes, its hedges or its reasons.
On Nasdaq, registered market makers are obliged to be willing to buy and sell continuously during regular hours, with quotes no further than a designated percentage from the national best bid or offer: 28% for most small caps priced at $1 or more and 30% below $1 under the current rule (checked 2026-10-05). That band is wide, so a market maker's quote can sit far from the inside. A market maker can be on the offer because the rule asks for a quote and its inventory is long, short or flat. A displayed quote is consistent with all three.
Displayed vs. executed liquidity
Level 2 is the displayed part of a larger set:
- Reserve and non-displayed orders sit on exchange books without showing their full size, or any size. Iceberg, reserve and hidden orders covers them.
- Off-exchange trades execute away from the lit book and are reported afterward. Off-exchange and dark pool prints covers how they appear on the tape.
- Cancelled orders vanish before they trade. Spoofing and disappearing walls covers what you can and cannot conclude from that.
- Venues your feed omits. A retail Level 2 package often shows a subset of venues, and some show only one exchange's book.
In the Marrowby example, 6,400 shares printed at $5.21 while MMQA showed 1,000. Some of those prints may have been MMQA's refreshes; others may have come from reserve orders at the same price or from a different venue entirely. The tape alone does not say.
"The ax": folklore vs. evidence
"The ax" is trader slang for the market maker said to control a stock's direction on a given day. The idea is appealing: find the big player, follow them. Three problems make it hard to use:
- Identity is not position. A firm can show an offer while buying elsewhere, or show a bid while its net position falls.
- The label is assigned after the move. Whoever was on the offer before a fade gets called the ax that day. Whoever was on the offer before a breakout is said to have "been run over". Both stories use the same observation.
- There is no published count. No source we found measures how often a refreshing MMID comes before a move in either direction. Without a count, the idea is a story, not a rate.
That does not make Level 2 useless. Size, refresh speed and where prints land relative to the quote are observations you can log and count. Names are the part that carries the least information.
Observe vs. infer
Go back to 09:47:30 in the Marrowby example (fictional; all figures Illustrative).
| What the screen shows (observed) | What a trader might infer | Why it is not established |
|---|---|---|
| MMQA shows 1,000 shares on the $5.21 offer | "MMQA is the seller" | A displayed quote shows one order, not the firm's net position |
| The $5.21 offer traded and reappeared six times since 09:46 | "MMQA is holding the stock down" | Refreshes fit a quoting obligation, inventory management or several sources at one price |
| 6,400 shares printed at $5.21 | "When MMQA leaves, it breaks out" | Past prints at a price say nothing about which side runs out first |
| The day's high so far is $5.24 | "The breakout level is $5.24" | A high is a label that holds until it does not |
Only the left column was usable at 09:47:30.
Odds box
The point is not that this situation loses. The point is that the identifier appears in all three paths, so it cannot by itself move the split. If a refreshing MMID carried real information, your own log would show a different split when it is present than when it is absent. That comparison needs a count.
What shifts the odds
These are hypotheses to record, not findings; no FennEdge sample tests them:
- Refresh count and speed. Six refreshes in 90 seconds vs. two in five minutes.
- Prints vs. display. Shares printed at the price divided by size displayed there.
- Where the other side sits. Thick bids one cent below vs. a gap of several cents.
- Whether the identifier is an exchange or a single firm. Exchange labels aggregate many traders.
- Time of day and halts. Quoting widens and steps away around LULD bands and halts.
Cost of being wrong
Using the Marrowby plan at 1,000 shares (Illustrative):
| Outcome | Exit | R after costs | Dollars after costs |
|---|---|---|---|
| Path A: target | $5.31 | +1.90R | +$95.00 |
| Path B: stop, 1¢ slippage | $5.15 | −1.30R | −$65.00 |
| Path C: time stop | $5.20 | −0.30R | −$15.00 |
| Path B, quotes pulled, 6¢ slippage | $5.10 | −2.30R | −$115.00 |
When market makers widen or step away, the book a stop relies on gets thinner exactly when the stop triggers. Size for the last row, not only for 1R.
How many trades before this means anything
To test "a refreshing MMID matters", you need two counts: outcomes when the pattern is present and when it is absent, under the same plan. With 20 cases in each group, the 95% intervals for a 30% vs. a 40% Path A rate overlap heavily (Wilson, about 15%–52% and 22%–61%; computed on hypothetical counts). How many trades before your numbers mean anything covers the method.
What this lesson does not claim
- It does not claim market makers manipulate or support any stock. A displayed quote is consistent with many motives.
- It does not claim Level 2 is useless. It claims the identifier is the least informative part of it.
- Marrowby Therapeutics and all identifiers are fictional; all figures are Illustrative.
Go deeper
- FennEdge first. FennEdge does not offer Level 2 depth data as of 2026-10-05; the Scanner (paid: Plus, Pro, Max) covers rapid rises and drops, regular-session highs and Nasdaq-listed LULD halts, without depth. The FennEdge article Level 2 vs. Time & Sales compares the two windows.
- Wholesaler internalization. Much retail flow executes off-exchange; see Off-exchange and dark pool prints.
- Quote stuffing vs. ordinary updates. A burst of quote changes is not by itself evidence of abuse; Spoofing covers the legal line.
- Routing. Where your own order goes changes what you see filled; see Order routing.
- Related: Tape reading, Thinking in probabilities.
Exercises
Six exercises follow this lesson: three concept checks, two replays of a static Level 2 snapshot (separate observation from inference; then pick every outcome consistent with the book at 09:47:30 before Path B is revealed), and one scenario that asks which reading of a refreshing offer is most defensible. Exercises, answers and progress are available after signing in.
Primary sources
- Nasdaq. Equity 2, Section 5: Market Maker Obligations.
- FINRA. Rule 6160.
- U.S. Securities and Exchange Commission. Proposed rescission of Rules 611 and 610(e), Release No. 34-105655 (June 11, 2026), describing off-exchange volume and retail order routing.
Sources checked 2026-10-05. Marrowby Therapeutics, all identifiers, quotes, prints, outcome splits and costs in this lesson are fictional and Illustrative. This lesson is education, not a recommendation to buy or sell any security. Past patterns do not indicate future results.