Build tomorrow’s watchlist the night before by screening for a catalyst, enough average volume to get filled, a daily range wide enough to pay for the risk, and price sitting near a level you have already marked. Three to five names is enough. Anything longer stops being a plan and becomes a feed.
The order of those filters matters as much as the filters themselves. Most people screen for movement first and end up with a list of stocks that already moved. Screening for why first leaves you with names that may still have somewhere to go.
The four filters, in the order they matter
1. Catalyst — why is this stock in play at all?
A stock that moves for no discoverable reason is a stock whose move you cannot anticipate. Earnings, guidance, a regulatory decision, an analyst action, a sector-wide move, an index change: any of these gives the day a reason to have participants in it. If you cannot write the catalyst in one short sentence, skip the name. “It was up a lot” is not a catalyst.
2. Liquidity — can you actually get in and out?
Average daily volume and the typical bid-ask spread decide whether your exit exists at the price you planned. A stock that trades 200,000 shares a day will move against you on the way out of a position you barely noticed putting on. This is the filter beginners skip and then blame on the setup, when the loss was really slippage.
3. Range — is there enough movement to pay for the risk?
If the average daily range is 1% and your stop needs 0.8% of room, the trade cannot pay for itself even when you are right. Compare the average true range against the stop distance your level actually requires. This is the same arithmetic the trading expectancy calculator runs, applied before the trade instead of after it.
4. Location — is price near something that matters?
The best names on a watchlist are ones sitting just under resistance, just above support, or at the edge of a range — because those are the places where a decision gets made and where an invalidation is cheap. A stock in the middle of nowhere may move tomorrow, but you will have no level to trade against. Mark the levels while the market is closed, using the same chart markup process every night.
Why the trending list is the worst starting point
Every broker and social platform publishes some version of a most-active or most-bought list, and it is tempting to treat it as a shortcut. It is not a shortcut, because it measures crowding rather than opportunity.
The evidence is unusually direct here. Studying Robinhood users, Barber, Huang, Odean and Schwarz reported average 20-day abnormal returns of −4.7% for the stocks most heavily purchased each day (Barber, Huang, Odean & Schwarz, “Attention-Induced Trading and Returns”, Journal of Finance, 2022). That is a study of holding periods far longer than a day trade, so it is not a rule about intraday behaviour. What it does establish is that a stock appearing on a most-bought list is evidence about where attention has already gone, not about where price is going.
Use those lists the way a scanner is meant to be used: as a source of candidates that then have to pass the four filters. A name that is trending and has a real catalyst, real liquidity, real range and a clean level is a good candidate. A name that is only trending is a crowd.
What a workable filter set looks like
| Filter | A reasonable starting threshold | Why it is there |
|---|---|---|
| Average daily volume | 1 million shares or more | You can enter and exit without being the reason price moved |
| Price | Above $5 | Avoids the widest spreads and the thinnest order books |
| Average true range | At least 2× your typical stop distance | The move has to be able to pay for the risk |
| Relative volume | Above 1.5× normal | Confirms today’s participation is unusual, not just the name |
| Catalyst | Writable in one sentence | Gives the session a reason to have a trend in it |
| Distance to level | Within about 1 ATR | The decision point is reachable in one session |
Treat those numbers as a starting point rather than a rule. A trader working a $2,000 account and a trader working a $200,000 account will set the volume filter very differently, because the size that moves the book is different. Adjust them from your own fills, not from someone else’s screenshot.
Three to five names, not thirty
Once the filters run, the temptation is to keep everything that passed. Resist it. The number of names you can genuinely track — holding the level, the invalidation and the size in your head while price is moving — is small, and it does not grow with experience nearly as much as people expect.
- Rank, then cut. Order the survivors by how clean the level is, not by how big the move was. Keep the top three to five.
- Prefer names you have traded before. A stock you know the rhythm of is worth more than a better-looking chart on a name you have never watched fill.
- Keep a second tier of two, unwatched. If a primary name gaps out of its level before the open, you have a replacement without improvising.
What to write down for each name
A watchlist that is only tickers is not preparation. For each name, write five things before you close the laptop:
- The catalyst, in one line.
- Two or three levels, with the one you would actually trade against marked.
- The direction you would take if price reaches each level — and explicitly, that you take nothing if it does not.
- The invalidation price, where the idea is simply wrong.
- The position size for that specific stop distance, worked out in advance with the position size calculator.
If you cannot write line four, the name comes off the list. An idea without an invalidation is a hope, and the argument for that is made in full in what invalidation actually means.
Your list is a hypothesis, not a commitment
Overnight news, a pre-market gap or a broad market move can invalidate the whole list before the bell, and the trader who feels obliged to use the work they did is the one who forces a trade. Re-check each name in the pre-market: if it has gapped clean through its level, the setup you prepared no longer exists, whatever the chart looked like at 10 p.m.
That re-check is a short, fixed routine rather than a fresh search, which is exactly what a pre-market routine is for. A no-trade day that follows a well-built list is a successful day, not a wasted one.
Frequently Asked Questions
How do you find stocks to day trade the night before?
Screen the day's movers for four things in order: a catalyst that explains why the stock is in play, average daily volume high enough that you can get filled without moving the price, an average daily range wide enough to pay for the risk you would take, and a price sitting near a level you have already marked on the chart. Names that pass all four go on the list. Everything else waits.
How many stocks should be on a day trading watchlist?
Three to five. The limit is attention, not opportunity. With five names you can hold each chart, each level and each invalidation in your head as the session moves. With thirty you will notice setups after they have already run, which is how chasing starts. A short list also makes your results readable, because you can tell which names actually suit your method.
Should you day trade the most popular or trending stocks?
Be careful. Popularity is a crowding signal, not an edge signal. Barber, Huang, Odean and Schwarz studied Robinhood users and reported average 20-day abnormal returns of negative 4.7 percent for the stocks most heavily bought each day. A trending list is a useful place to see where attention is, but it is a poor place to choose what to trade, because by the time a name is on it the easy move has usually happened.
What should you write down for each stock on your watchlist?
The catalyst in one line, two or three price levels with the level you would trade against marked clearly, the direction you would take if price reaches each level, the invalidation price where the idea is wrong, and your position size for that specific stop distance. If you cannot write the invalidation, the idea is not ready and the name should come off the list.
Bottom line
Selection is the part of equity day trading that happens while nothing is at stake, which is exactly why it is the part worth systematising. Four filters, three to five names, five written lines each, and a pre-market re-check. Do that consistently and most of the bad trades you would otherwise take never get the chance to exist, because they were never on the list. The wider context sits in day trading stocks: what is different about equities, and the confirmation step at the open is covered in volume analysis.