Explainer · Kresmion Research
What Is Paper Trading? What It Teaches, What It Cannot, and How Club Competitions Score It
Published by Kresmion Research. Read our editorial approach and data methodology.
Paper trading is placing simulated trades with virtual money at real market prices, so a strategy or a platform can be practised with no real cash at risk.
Brokers, charting tools, classrooms and university finance clubs offer versions of it, from a practice account that mirrors a live one to a ranked competition with a season and published rules. How much a paper record means depends on how honestly the simulator fills orders and charges costs. This page covers what paper trading is, what it can and cannot teach, how fills and costs should be modelled for the results to mean something, how paper-trading competitions are scored, and how the competitions we run for university finance clubs work. It is descriptive throughout.
What paper trading is
A paper account starts with a sum of virtual cash. Orders placed in it are matched against real prices, live or delayed, and the account tracks positions, cash and profit or loss as if the trades had happened, but no order ever reaches a market.
It is different from backtesting. A backtest replays fixed rules over prices that already exist, which the person building it may already have seen; a paper account runs forward on prices that have not happened yet. That makes it slower, and closer to the experience of actually trading: the next price is unknown when the order goes in.
What it can teach
- Order mechanics. How a market order, a limit order and a stop order behave, and why they fill differently; market order vs limit order and what a stop order is cover each. The difference between the price on the screen and the price received shows up on the first trade in a simulator that fills realistically.
- Sizing and exposure. How much of an account one position takes, how position size follows from a chosen risk, and what borrowing to buy or short selling does to the account's swings.
- Record keeping. A paper account keeps a full log of every order, fill and fee, which turns vague recollection into something that can be reviewed.
- Measuring results. Return alone says little; the size of the swings taken to get it, the drawdown along the way and the Sharpe ratio put it in context.
- Routine. Following a written process every session for weeks is a habit, and it can be practised without money at risk.
What it cannot teach
- Emotion. Nothing is at stake, so a paper loss does not hurt the way a real one does, and the decisions a trader makes under that pressure go untested.
- Real fills. A paper order does not join a real order book. The SEC's investor bulletin on order types notes that the price at which a market order will be executed is not guaranteed, that the last-traded price is not necessarily the price at which a market order will be executed, and that in a fast-moving market parts of a large market order can execute at different prices. Its example is a market order that may execute at a higher price if other orders are executed first. A limit order, the same bulletin says, is not guaranteed to execute and can only be filled if the market price reaches the limit price. Even then, reaching the price is not enough if other orders at that price sit ahead in the queue.
- Market impact. Paper money cannot move a price. A real buy order larger than the shares offered at the best price fills the rest at higher prices, while a paper order never uses up the shares on offer.
- Every cost. Spreads, commissions, borrow fees on shorts, interest on borrowed cash and taxes all come out of a real account. A simulator that skips them flatters every strategy that trades often.
- Proof of skill, over a short window. A few weeks of daily returns is a small sample, and one large winning bet can dominate it.
How fills and costs should be modelled
A paper result is only as honest as the rules behind it. A simulator that aims to be realistic does the following:
1. Fills after the order, never before. The fill price comes from a price recorded after the order was placed, not the price on the screen when it was clicked, so nobody can trade on a price they have already seen move. 2. Charges the spread and slippage. A buy pays more than the reference price and a sale receives less, as on a real market where buyers pay the ask and sellers receive the bid; see the bid-ask spread. 3. Charges commission. Every fill pays a fee. 4. Fills limits only through the price. A limit order fills when the market trades through its price, not when it merely touches it, which stands in for the queue. 5. Caps size against volume. Since paper orders cannot move the price, a cap on order size as a share of the instrument's typical daily trading volume stands in for market impact. 6. Respects market hours. An order placed while the market is closed fills at the next session's opening price, not at the last close. 7. Applies corporate actions and carrying costs. Splits, dividends, borrow fees on shorts and interest on borrowed cash all reach the account.
A hypothetical example, using the cost schedule of Kresmion's club competitions for a stock whose median daily trading over the last 20 sessions is between $20 million and $100 million (slippage 5 basis points, commission 0.05 percent of the trade's value). A member buys 200 shares at 10:12:37 New York time while the screen shows a last price of $50.00:
| Step | Value |
|---|---|
| Fill bar | The 1-minute bar that starts at 10:13:00, opening at $50.10 |
| Fill price after 5 bp slippage | $50.10 x 1.0005 = $50.12505 |
| Trade value | 200 x $50.12505 = $10,025.01 |
| Commission (0.05%) | $5.01 |
| Cash paid | $10,030.02 |
| Naive cost at the screen price | 200 x $50.00 = $10,000.00 |
The realistic fill costs $30.02 more than the naive one: $20.00 from the price moving to the next bar's open, $5.01 of slippage and $5.01 of commission. The next bar could as easily have opened lower; the costs are charged either way. If the member later sells when both the screen and the next bar's open show $51.00, the sale fills at $50.9745 and pays about $5.10 of commission, for net proceeds of $10,189.80. The profit is $159.78, against the $200.00 a simulator with no costs and instant fills would have reported.
How paper-trading competitions are scored
A competition gives every entrant the same starting cash, the same window and the same rules, then ranks them. Ranked on return alone, a short contest rewards the entrant who put everything on one volatile bet and happened to be right, which says more about luck than about process. A competition can use several tools against that, and ours use all four:
- Risk-adjusted scoring, which divides return by volatility, in the spirit of the Sharpe ratio.
- Eligibility rules, such as a minimum number of sessions since the first trade, a minimum level of investment or a minimum number of asset types held.
- Position caps and a freeze on new risk in the final sessions, so the last days cannot be used for an all-or-nothing swing.
- Rules published and locked before the start, so nobody can change them mid-season.
A hypothetical pair over one season, in a contest with no position caps or eligibility gates, shows why the score matters. Member A holds one volatile stock, with a mean daily return of 0.30 percent and a daily standard deviation of 3.0 percent. Member B holds a spread of assets, with a mean daily return of 0.08 percent and a daily standard deviation of 0.5 percent. A has the higher return. Annualised, A's volatility is 47.6 percent and B's is 7.9 percent, so on the formula in the next section A scores 0.756 / 0.476 = 1.59 and B scores 0.202 / 0.10 = 2.02 (B's volatility is below the 10 percent floor, so the floor is used). B ranks first on the risk-adjusted score and A on return. In a Kresmion season on the default rules, A could not build that book at all: one stock would break the 25 percent position cap, and a single position held at the cap would also miss the asset-type and average-invested gates below, so A would not be ranked.
How Kresmion's club competitions work
Kresmion gives university finance clubs a space of their own, and the club's board can run paper-trading seasons in it, described on the for clubs page. Members trade virtual cash at real market prices; no real money is involved and no order reaches a market. The rules below are the defaults and limits shown on that page and in the season rules, as read on 7 November 2026; a board sets its own season within them, and the rules lock when the season opens.
| Rule | Default, and what the board can change |
|---|---|
| Length | 4 weeks by default; the board picks any span from 1 to 16 weeks |
| Starting cash | 100,000 by default (the board sets 10,000 to 1,000,000), in EUR or USD |
| What members can trade | Thousands of stocks and, with the wider list on (the default), hundreds of ETFs and dozens of cryptocurrencies, plus the G10 currencies |
| Orders | Market, limit and stop orders and short selling, by default (the board can switch off limit and stop orders, and shorting); cash dividends credited to longs and charged to shorts |
| Exposure cap | Gross exposure up to 150 percent of the account by default; the board sets 100 to 200 percent |
| Position limits | By default no position above 25 percent of the account and no asset type above 60 percent at the time of an order, and no new risk in the final 2 sessions; the board can change each |
| Fills | Fixed: the next 1-minute bar during market hours, the next official open outside them; crypto at the next 1-minute exchange price; FX at the next ECB reference rate |
| Costs | Fixed: commission of 0.05 percent on stocks and ETFs (at least 1 unit of the base currency), 0.10 percent on crypto and 0.02 percent on FX, plus slippage of 1 to 30 basis points depending on how much the instrument trades |
| Size cap | Fixed: no order above 1 percent of a stock's typical daily trading (0.1 percent for crypto) |
| Carrying costs | Borrowed cash costs 5 percent a year by default (the board sets 0 to 10 percent); a short pays a fixed borrow fee of 0.5 percent a year |
The headline ranking is the risk-adjusted score or plain return, as the board chooses (risk-adjusted by default), and both are shown. The page states the score as:
score = mean(r) × 252 ÷ max( sd(r) × √252 , 0.10 )
where r is the account's daily return, the percentage change in its value from one session to the next. It is the annualised mean daily return divided by the annualised volatility, a Sharpe-style ratio with no risk-free rate, and the 10 percent floor on volatility stops a nearly idle account from posting a large score on tiny swings. To be ranked, a member must:
- be at least 5 sessions in, counting the session of their first fill (no further trade is needed);
- hold the minimum number of asset buckets the board sets, 3 by default (a bucket counts once held at 5% or more for 3 sessions);
- keep average gross exposure at 50% or more since their first fill;
- hold positions on at least 70% of the season's sessions.
Ties go to the lower maximum drawdown, then the earlier first fill. Members who miss a rule are listed under the ranked ones with the rule they missed, and standings update each morning at 07:00 UTC. As the page says, paper rankings measure outcomes in a simulation and are not investment advice.
Honest limitations
The worked examples are hypothetical. The Kresmion rules above are the defaults and limits as of 7 November 2026; a club board sets its own season within them, and a season keeps the rules it opened with. Paper fills use recorded prices with modelled costs, so a real order of the same size could fill at a different price or not at all, and prices come from market data providers and may be delayed. A four-week season holds about 20 trading sessions, a short record for any score, risk-adjusted or not. Kresmion is not a broker: it does not hold client money or route orders.
Key takeaways
| Point | Detail |
|---|---|
| Definition | Simulated trades with virtual money at real market prices; no order reaches a market |
| What it teaches | Order mechanics, sizing, record keeping, measuring results, routine |
| What it cannot | Emotion under real losses, real fills and queues, market impact, every real cost, proof of skill over a few weeks |
| Realistic modelling | Fill after the order, charge spread, slippage and commission, fill limits only through the price, cap size against volume |
| Competitions | Risk-adjusted scores, eligibility rules, position caps and locked rules make it harder for one lucky bet to decide the ranking |
| Kresmion clubs | score = mean(r) × 252 ÷ max( sd(r) × √252 , 0.10 ) on daily returns, four ranking gates, next 1-minute bar fills, exposure cap of 150 percent by default (100 to 200) |
Frequently asked questions
What is paper trading used for?
It is used to learn how orders, fills and accounts work, to test a trading process forward in time without risking money, and to run classroom and club competitions. It records simulated trades against real prices, so the mechanics are real while the money is not.
Is paper trading the same as backtesting?
No. A backtest replays fixed rules over prices that already exist, which the person building it may already have seen, while paper trading makes one decision at a time on prices that have not happened yet. The two answer different questions and can be used together.
Why do paper trading results differ from real trading?
A paper order does not join a real order book, cannot move the price and, in a simple simulator, may pay no spread, slippage or commission. Real market orders are not guaranteed a price and limit orders are not guaranteed to fill, so a realistic simulator fills after the order, charges costs and caps order size against trading volume.
How are paper trading competitions scored?
Ranked on return alone, a short contest rewards concentrated bets. Kresmion's club competitions offer a risk-adjusted score, score = mean(r) × 252 ÷ max( sd(r) × √252 , 0.10 ), where r is the account's daily percentage return, alongside plain return, with eligibility gates on sessions since the first trade, asset types held, average exposure and activity.
Does a high paper-trading ranking show someone will trade well with real money?
No. A paper ranking measures outcomes in a simulation over a short window, without real fills, market impact or the pressure of real losses. It describes how a process performed under the competition's rules, not how it will perform with money at risk.
This page is information, not investment advice.
---
Source: SEC Office of Investor Education and Advocacy, Investor Bulletin "Trading Basics: Understanding the Different Ways to Buy and Sell Stock" (market and limit order execution), https://www.sec.gov/files/trading101basics.pdf ; Kresmion for clubs page (season defaults, exposure cap, fills, score formula, ranking gates, ties), https://kresmion.com/for-clubs, read 7 November 2026 ; Kresmion paper-trading season rules card (costs, size cap, position limits, carrying costs, standings time), as in force for seasons created from 3 November 2026 ; worked examples: hypothetical figures computed by Kresmion.
Kresmion Research.
- · SEC Office of Investor Education and Advocacy, Investor Bulletin, Trading Basics: Understanding the Different Ways to Buy and Sell Stock: https://www.sec.gov/files/trading101basics.pdf
- · Kresmion for clubs page (season defaults, exposure cap, fills, score formula, ranking gates, ties), read 7 November 2026: https://kresmion.com/for-clubs
- · Kresmion paper-trading season rules card (costs, size cap, position limits, carrying costs, standings time), seasons created from 3 November 2026
- · Worked fill, cost and scoring examples: hypothetical figures computed by Kresmion
Put this to work
Real filings, 13F flows, and positioning reads with their sources, in your inbox when there is something in the data, or live on Telegram. Free, no account.
No fixed schedule: it goes out when the data has something in it. Unsubscribe anytime.
Kresmion publishes information, not investment advice. See our methodology and the latest research notes.