The short answer
Risk management for a day trader is four numbers decided before the open and one piece of arithmetic that follows from them. R is the most you lose on one trade when the stop is hit, a fixed fraction of the account. Position size is R divided by the distance from entry to stop, so the stop sets the size and never the reverse. The daily loss limit is two or three R, and the day ends when it is spent. The maximum number of trades is small enough that you can remember each one. Everything else is drawdown arithmetic: a 20% loss needs a 25% gain to recover, a 50% loss needs 100%, and a losing streak that is certain to arrive must be survivable at the size you chose. Below are the definitions and the worked examples, with numbers that check.
What is R, and why not just use dollars?
R is one unit of risk: the dollar amount you lose if a trade goes to its stop. Decide it as a fraction of the account. On a $25,000 account at 0.5%, R is $125. Every trade is then measured in R. A trade that makes $250 is +2R. A trade stopped out is -1R. A trade that was supposed to lose $125 and lost $400 because the stop was moved is -3.2R, and that number is the whole story of the day.
Dollars hide what R shows. Two traders can both be up $600 on the week; one made it in three trades at +1.6R each and the other in eleven trades that averaged +0.4R with a -4R day in the middle. Only R makes the second record visible. It also makes records comparable across account sizes and instruments, which is why journals use it.
R only means something if the stop is real. If the stop moves, the denominator moves, and nothing after it means anything. It also has to be accepted in the first place: why a broker rejects a stop on the wrong side of the price is its own short guide, written for Robinhood.
An illustration. The prices are invented.
How much should I risk per trade?
A fixed, small fraction of the account, chosen so that a normal losing streak is a bad week rather than a closed account. Common practice sits between half a percent and one percent. The argument for the smaller end is arithmetic, shown below, and it applies most to anyone who has not yet got a hundred trades of record.
The number is not the point. The fixedness is. Risk per trade that floats with confidence is the mechanism by which one wrong trade at four times size undoes eight right ones.
How do I size a position from the stop distance?
Shares or contracts = R divided by the loss per share or contract between entry and stop. The position size calculator does this division in your browser, starting from the day's budget; here it is by hand, with R = $125.
Options first, because an option's loss per contract is the premium distance times 100, the contract multiplier on SPY and SPX options. Buy one SPY call at $2.50, which is $250, with a stop at $1.50: the distance is $1.00, the loss per contract is $100, and 125 / 100 = 1.25, round down to 1 contract risking $100. Options stops fill badly, so the honest worst case is the whole premium, $250, which is 2R. Size for the honest worst case.
Stocks work the same way without the multiplier. Three worked examples:
- Stock, entry $50.00, stop $49.50. Distance $0.50. 125 / 0.50 = 250 shares. Position value $12,500.
- Stock, entry $120, stop $118. Distance $2. 125 / 2 = 62.5, round down to 62 shares. Risk $124. Position value $7,440.
- Stock, entry $50.00, stop $49.75. Distance $0.25. 125 / 0.25 = 500 shares. Position value $25,000, the whole account. A tight stop buys a big position, and a big position pays more for slippage and needs margin. The tight stop is not free.
Futures follow the same rule with the contract multiplier. On a micro E-mini S&P contract, one point is $5. An eight-point stop risks $40 a contract, so 125 / 40 = 3.125, round down to 3 contracts risking $120. On the full-size contract a point is $50, an eight-point stop is $400, and 125 / 400 is zero contracts. The answer "not this instrument at this account size" is a valid output of the formula and the one people most often refuse to accept.
How many contracts should I trade?
As many as R buys at your stop, rounded down, and none if that is less than one. For an SPX or SPY option the loss per contract is the premium distance times 100. With R at $125 and a stop $0.40 under the entry, a contract risks $40, and 125 / 40 = 3.125: three contracts.
Then check the whole premium. A stop on an option is a price, not a promise, and a 0DTE option can go through it, or to zero, the same afternoon. Three contracts bought at $2.50 are $750, which is 6R. If that is more than the day's limit, the answer is fewer contracts or a cheaper strike.
The number does not grow after a win or shrink after a loss. It changes when R does. The free position size calculator returns the size, how many losing trades it buys, and where the day ends.
What should my daily loss limit be?
Two or three R. With R at $125 that is $250 to $375, or 1% to 1.5% of the account. The test is not whether you can afford it once. It is whether you could hit it four days running and turn up on the fifth sizing exactly the same way. If one losing trade would spend the whole day, the size is wrong, not the limit. What Interactive Brokers itself offers for a limit, and what it does not, is in how to set a daily loss limit on Interactive Brokers.
Why a limit at all, if each trade is already capped? Because the trade after two losses is not taken to catch a setup, it is taken to get the two back, and it is sized to make that arithmetic work. Coval and Shumway measured this in professionals: Chicago Board of Trade traders who lost in the morning were more likely to take above-average risk in the afternoon than traders who had gained, 31.2% against 27%, and they placed more and larger trades. Here is what that looks like in R. Lose 1R, double, lose 2R, double, lose 4R: 7R in three trades, 3.5% of the account at R = 0.5%. The same three losses at fixed size cost 3R, and a 3R limit would have ended the day before the third one.
An illustration. The days are invented.
How many trades a day before it is overtrading?
Few enough to remember each one at the review. For most people that is somewhere between two and six, and past the number you set in the morning it is overtrading. Every trade is a chance to break a rule, a commission and a spread paid against an edge that was small to begin with, and a thing you will have to write up. Sixty trades a day is sixty chances and no memory of any of them. Scale, when it comes, comes from size on a setup the record has earned, not from frequency. Finding your own number, and holding it, is in how to stop overtrading.
The drawdown arithmetic
The gain needed to recover a loss is the loss divided by what is left: L / (1 - L).
| Drawdown | Gain needed to recover |
|---|---|
| 5% | 5.3% |
| 10% | 11.1% |
| 20% | 25% |
| 30% | 42.9% |
| 50% | 100% |
| 70% | 233% |
Check one: lose 30% of $10,000 and you have $7,000; to get back you need $3,000, which is 42.9% of $7,000. Check two: lose half of $10,000 and you have $5,000, which has to double, a 100% gain, to get back. The curve is gentle at the top and vertical at the bottom, which is the entire argument for keeping the daily limit small. If you are already at the bottom of it, what to do in the week after blowing up an account comes first.
Now the two settings compared over a bad stretch. Ten limit days in a row, nothing else:
- Limit at 1.5% a day: 0.985 to the tenth power is 0.860, a 14% drawdown, needing 16% to recover.
- Limit at 3% a day: 0.97 to the tenth power is 0.737, a 26% drawdown, needing 36% to recover.
Ten straight limit days is rare. Five is not, over a year. Five at 1.5% is a 7.3% drawdown; five at 3% is 14.1%.
How likely is a losing streak?
Certain, given enough trades. With a 45% win rate, the chance of any five consecutive trades all losing is 0.55 to the fifth power, about 5%. In a hundred trades there are nearly a hundred overlapping windows of five, so you will see a five-loss streak, and probably a six. At R = 0.5% and fixed size, five in a row is 2.5% of the account and a bad week. At size that floats up after losses, the same five trades can be the account.
The other side of the same arithmetic is expectancy: what a trade is worth on average over many repetitions. Take invented numbers: 45% winners averaging +1.8R and 55% losers at -1R. Expectancy is 0.45 × 1.8 minus 0.55 × 1, which is +0.26R per trade, about $32 at R = $125, before costs. That is an average over hundreds of trades, not a Tuesday, and it only exists if the -1R losses are actually -1R. It is also invented; your record is the only source of your numbers, and it needs something like a hundred trades before it says anything.
Why the base rates matter here
Barber and Odean's study of 66,465 US households found the most active traders earned 11.4% a year while the market returned 17.9%; the activity itself was the cost. In Brazil, 97% of those who persisted at day trading for more than 300 days lost money. In Taiwan, less than 1% of day traders reliably earned positive returns net of fees. None of those studies checked whether the losers had a stop they could not move or a limit they could not override, so none of them proves that the four numbers above fix anything. What the numbers do is keep the account alive long enough to find out which group you are in.
Where Degen Pro fits
Degen Pro is a Mac app for Interactive Brokers whose journal is written in R. Position sizing works the size out from the stop and the budget, not from the last trade, so the division above is never done in the moment it matters, and it says when the record supports sizing up and when it does not. The daily budget per instrument, the trade count and the stand-down after a run of reds are set before the open and frozen once armed; on the trade after the limit the app counts it, warns you, ends the day, or ends the day and locks Trader Workstation and TradingView, whichever you chose that morning. If you trade a prop account, its floor, daily limit, target and consistency rule are tracked beside your own numbers, with every attempt and its fee. The weekly card shows rules kept and a mark for every trade, and no dollar figure and no R; the R stays in the app. It gives no signals — it can place an entry you choose, calls and puts on SPX and SPY, but it never tells you what to take; Mac only, and the lock needs Interactive Brokers.
The four numbers, once more
- R: a fixed fraction of the account, the same every trade.
- Size: R divided by the stop distance, rounded down.
- Daily limit: two or three R, and the day ends there.
- Trades per day: few enough to review.
Decide them in the morning. The arithmetic does the rest, and it does not care how you feel at noon. None of this is investment advice; trading involves risk of loss, and the examples above are illustrations with made-up numbers, not forecasts.